In [1]:
# Computations
import numpy as np
import pandas as pd
# scipy
import scipy.stats as stats
# sklearn
from sklearn.metrics import classification_report, accuracy_score, f1_score, precision_score, recall_score
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split, GridSearchCV, RandomizedSearchCV, cross_val_score
# keras
import keras
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.optimizers import SGD
from keras.utils.vis_utils import plot_model
import keras.backend as K
# Visualisation libraries
import seaborn as sns
import matplotlib.pyplot as plt
from matplotlib.patches import Ellipse, Polygon
import matplotlib.gridspec as gridspec
import missingno as msno
import plotly.offline as py
import plotly.express as px
import plotly.figure_factory as ff
import plotly.graph_objs as go
from plotly.offline import init_notebook_mode, iplot
from plotly.subplots import make_subplots
from wordcloud import WordCloud
import re
# Graphics in retina format
%config InlineBackend.figure_format = 'retina'
# sns setting
sns.set_context("paper", rc={"font.size":12,"axes.titlesize":14,"axes.labelsize":12})
sns.set_style("whitegrid")
# plt setting
plt.style.use('seaborn-whitegrid')
plt.rcParams['axes.labelsize'] = 14
plt.rcParams['xtick.labelsize'] = 12
plt.rcParams['ytick.labelsize'] = 12
plt.rcParams['text.color'] = 'k'
%matplotlib inline
import warnings
warnings.filterwarnings("ignore")
Using TensorFlow backend.
Pima Indians Diabetes Data Classification using Feature Importance¶
In this article, we use Kaggle'sPima Indians Diabetes. The Pima indians are a group of Native Americans living in an area consisting of what is now central and southern Arizona. A variety of statistical methods are used here for predictions.
Context¶
This dataset is originally from the National Institute of Diabetes and Digestive and Kidney Diseases. The objective of the dataset is to diagnostically predict whether or not a patient has diabetes, based on certain diagnostic measurements included in the dataset. Several constraints were placed on the selection of these instances from a larger database. In particular, all patients here are females at least 21 years old of Pima Indian heritage.
Content¶
The datasets consist of several medical predictor variables and one target variable, Outcome. Predictor variables include the number of pregnancies the patient has had, their BMI, insulin level, age, and so on.
Citations¶
Table of contents¶
Dataset Analysis¶
In [2]:
Data = pd.read_csv('pima-indians-diabetes-database/diabetes.csv')
display(Data.head())
print('The Dataset Shape: %i rows and %i columns' % Data.shape)
| Pregnancies | Glucose | BloodPressure | SkinThickness | Insulin | BMI | DiabetesPedigreeFunction | Age | Outcome | |
|---|---|---|---|---|---|---|---|---|---|
| 0 | 6 | 148 | 72 | 35 | 0 | 33.6 | 0.627 | 50 | 1 |
| 1 | 1 | 85 | 66 | 29 | 0 | 26.6 | 0.351 | 31 | 0 |
| 2 | 8 | 183 | 64 | 0 | 0 | 23.3 | 0.672 | 32 | 1 |
| 3 | 1 | 89 | 66 | 23 | 94 | 28.1 | 0.167 | 21 | 0 |
| 4 | 0 | 137 | 40 | 35 | 168 | 43.1 | 2.288 | 33 | 1 |
The Dataset Shape: 768 rows and 9 columns
| Feature | Explanations |
|---|---|
| Pregnancies | Number of times pregnant |
| Glucose | Plasma glucose concentration a 2 hours in an oral glucose tolerance test |
| BloodPressure | Diastolic blood pressure (mm Hg) |
| SkinThickness | Triceps skinfold thickness (mm) |
| Insulin | 2-Hour serum insulin (mu U/ml) |
| BMI | Body mass index (weight in kg/(height in m)^2) |
| DiabetesPedigreeFunction | Diabetes pedigree function |
| Age | Age (years) |
| Outcome | Whether or not a patient has diabetes |
In [3]:
def Data_info(Inp, Only_NaN = False):
Out = pd.DataFrame(Inp.dtypes,columns=['Data Type']).sort_values(by=['Data Type'])
Out = Out.join(pd.DataFrame(Inp.isnull().sum(), columns=['Number of NaN Values']), how='outer')
Out['Percentage'] = np.round(100*(Out['Number of NaN Values']/Inp.shape[0]),2)
if Only_NaN:
Out = Out.loc[Out['Number of NaN Values']>0]
return Out
display(Data_info(Data).T[:2])
_ = msno.bar(Data, figsize=(12,3), fontsize=14, log=False, color="#34495e")
display(Data.describe())
| Age | BMI | BloodPressure | DiabetesPedigreeFunction | Glucose | Insulin | Outcome | Pregnancies | SkinThickness | |
|---|---|---|---|---|---|---|---|---|---|
| Data Type | int64 | float64 | int64 | float64 | int64 | int64 | int64 | int64 | int64 |
| Number of NaN Values | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Pregnancies | Glucose | BloodPressure | SkinThickness | Insulin | BMI | DiabetesPedigreeFunction | Age | Outcome | |
|---|---|---|---|---|---|---|---|---|---|
| count | 768.000000 | 768.000000 | 768.000000 | 768.000000 | 768.000000 | 768.000000 | 768.000000 | 768.000000 | 768.000000 |
| mean | 3.845052 | 120.894531 | 69.105469 | 20.536458 | 79.799479 | 31.992578 | 0.471876 | 33.240885 | 0.348958 |
| std | 3.369578 | 31.972618 | 19.355807 | 15.952218 | 115.244002 | 7.884160 | 0.331329 | 11.760232 | 0.476951 |
| min | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.078000 | 21.000000 | 0.000000 |
| 25% | 1.000000 | 99.000000 | 62.000000 | 0.000000 | 0.000000 | 27.300000 | 0.243750 | 24.000000 | 0.000000 |
| 50% | 3.000000 | 117.000000 | 72.000000 | 23.000000 | 30.500000 | 32.000000 | 0.372500 | 29.000000 | 0.000000 |
| 75% | 6.000000 | 140.250000 | 80.000000 | 32.000000 | 127.250000 | 36.600000 | 0.626250 | 41.000000 | 1.000000 |
| max | 17.000000 | 199.000000 | 122.000000 | 99.000000 | 846.000000 | 67.100000 | 2.420000 | 81.000000 | 1.000000 |
Let's take a close look at our data.
In [4]:
fig, ax = plt.subplots(nrows=4, ncols=2, figsize = (16, 20))
for i in range(len(Data.columns[:-1])):
sns.distplot(Data.iloc[:,i], rug=True, rug_kws={"color": "red"},
kde_kws={"color": "k", "lw": 2, "label": "KDE"},
hist_kws={"histtype": "step", "linewidth": 2,
"alpha": 1, "color": "Navy"}, ax= ax[int(i/2),i%2])
if Data.iloc[:,i].name != 'BMI':
ax[int(i/2),i%2].set_xlabel(re.sub(r"(\w)([A-Z])", r"\1 \2", Data.iloc[:,i].name))
In [5]:
Temp = ['Non-Diabetic' if x==0 else 'Diabetic' for x in Data['Outcome']]
fig = go.Figure(data=go.Splom(dimensions=[dict(label='Pregnancies', values=Data['Pregnancies']),
dict(label='Glucose', values=Data['Glucose']),
dict(label='Blood<br>Pressure', values=Data['BloodPressure']),
dict(label='Skin<br>Thickness', values=Data['SkinThickness']),
dict(label='Insulin', values=Data['Insulin']),
dict(label='BMI', values=Data['BMI']),
dict(label='Diabetes<br>Pedigree<br>Fun', values=Data['DiabetesPedigreeFunction']),
dict(label='Age', values=Data['Age'])],
showupperhalf=False,
marker=dict(color=Data['Outcome'], size=4, colorscale='Bluered',
line=dict(width=0.4, color='black')),
text=Temp, diagonal=dict(visible=False)))
del Temp
fig.update_layout(title='Scatterplot Matrix', dragmode='select',
width=900, height=900, hovermode='closest')
fig.show()
As can be seen, the Data has a normal distribution, and some entries need to be adjusted. In doing so, we defined a normalizer as follows, for a given vector $x$,
\begin{align*} \text{Normalizer}(x, cut) = \begin{cases} x_i &\mbox{if } |x_i- \mu|<\sigma\times cut \\ mode(x) & \mbox{else} \end{cases}. \end{align*}In [6]:
def Normalizer(Col, cut = 3):
return Col[(Col > (Col.mean() - Col.std() * cut)) &
(Col < (Col.mean() + Col.std() * cut))]
fig, ax = plt.subplots(nrows=4, ncols=2, figsize = (16, 20))
Temp = Data.copy()
for i in range(len(Data.columns[:-1])):
Data[Data.columns[i]] = Normalizer(Data[Data.columns[i]])
Data[Data.columns[i]] = Data[Data.columns[i]].fillna(Data[Data.columns[i]].dropna().mode()[0])
# Sub-Plots
sns.distplot(Data.iloc[:,i], rug=True, rug_kws={"color": "red"},
kde_kws={"color": "k", "lw": 2, "label": "KDE"},
hist_kws={"histtype": "step", "linewidth": 2,
"alpha": 1, "color": "Navy"}, ax= ax[int(i/2),i%2])
if Data.iloc[:,i].name != 'BMI':
ax[int(i/2),i%2].set_xlabel(re.sub(r"(\w)([A-Z])", r"\1 \2", Data.iloc[:,i].name))
Basically, we diminished the influence of certain data points (see the following figure).
In [7]:
Temp0 = Temp.copy()
Temp0.iloc[:,:-1] = abs(Data.iloc[:,:-1] - Temp.iloc[:,:-1])
Temp = ['Non-Diabetic' if x==0 else 'Diabetic' for x in Temp0['Outcome']]
fig = go.Figure(data=go.Splom(dimensions=[dict(label='Pregnancies', values=Temp0['Pregnancies']),
dict(label='Glucose', values=Temp0['Glucose']),
dict(label='Blood<br>Pressure', values=Temp0['BloodPressure']),
dict(label='Skin<br>Thickness', values=Temp0['SkinThickness']),
dict(label='Insulin', values=Temp0['Insulin']),
dict(label='BMI', values=Temp0['BMI']),
dict(label='Diabetes<br>Pedigree<br>Fun', values=Temp0['DiabetesPedigreeFunction']),
dict(label='Age', values=Temp0['Age'])],
showupperhalf=False,
marker=dict(color=Temp0['Outcome'], size=4, colorscale='Bluered',
line=dict(width=0.4, color='black')),
text=Temp, diagonal=dict(visible=False)))
del Temp, Temp0
fig.update_layout(title='Scatterplot Matrix', dragmode='select',
width=900, height=900, hovermode='closest')
fig.show()
Data Correlation¶
In [8]:
def Correlation_Plot (Df,Fig_Size):
Correlation_Matrix = Df.corr()
mask = np.zeros_like(Correlation_Matrix)
mask[np.triu_indices_from(mask)] = True
for i in range(len(mask)):
mask[i,i]=0
Fig, ax = plt.subplots(figsize=(Fig_Size,Fig_Size))
sns.heatmap(Correlation_Matrix, ax=ax, mask=mask, annot=True, square=True,
cmap =sns.color_palette("RdYlGn", n_colors=10), linewidths = 0.2, vmin=0, vmax=1, cbar_kws={"shrink": .7})
bottom, top = ax.get_ylim()
Correlation_Plot (Data, 9)
In [9]:
Temp = Data.iloc[:,:-1].var().sort_values(ascending = False).to_frame(name= 'Variance')
display(Temp)
Temp0 = Data.corr()
Temp0.loc[Temp.index[-1]].sort_values().to_frame(name= 'Correlation')[:-1].T
| Variance | |
|---|---|
| Insulin | 7844.510917 |
| Glucose | 929.680350 |
| SkinThickness | 246.979708 |
| BloodPressure | 146.573540 |
| Age | 128.991301 |
| BMI | 43.941176 |
| Pregnancies | 10.734190 |
| DiabetesPedigreeFunction | 0.078702 |
Out[9]:
| Pregnancies | BloodPressure | Age | Glucose | BMI | SkinThickness | Insulin | Outcome | |
|---|---|---|---|---|---|---|---|---|
| Correlation | 0.015703 | 0.034428 | 0.066525 | 0.095686 | 0.122868 | 0.15229 | 0.184028 | 0.192156 |
Even though the variance of Diabetes Pedigree Function is low, this might not improve the performance of the model, the correlation of this feature with the reset of features, especially with the Outcome, is noticeable.
Modeling and Classification¶
In [10]:
Target = 'Outcome'
X = Data.drop(columns = [Target])
y = Data[Target]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
pd.DataFrame(data={'Set':['X_train','X_test','y_train','y_test'],
'Shape':[X_train.shape, X_test.shape, y_train.shape, y_test.shape]}).set_index('Set').T
Out[10]:
| Set | X_train | X_test | y_train | y_test |
|---|---|---|---|---|
| Shape | (537, 8) | (231, 8) | (537,) | (231,) |
Furthermore, we would like to standardize features by removing the mean and scaling to unit variance.
In [11]:
scaler = StandardScaler()
X_train_STD = scaler.fit_transform(X_train)
X_test_STD = scaler.transform(X_test)
X_train_STD = pd.DataFrame(data = X_train_STD, columns = X_train.columns)
X_test_STD = pd.DataFrame(data = X_test_STD, columns = X_test.columns)
Here, we implement an artificial neural network (ANN) using Keras.
In [12]:
model = Sequential()
model.add(Dense(12, input_dim= X_train_STD.shape[1], init='uniform', activation='relu'))
model.add(Dense(10, init='uniform', activation='relu'))
model.add(Dense(1, init='uniform', activation='relu'))
# Number of iterations
N = int(1e3)
def mean_pred(y_true, y_pred):
return K.mean(y_pred)
model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy', mean_pred])
# Train model
history = model.fit(X_train_STD, y_train, nb_epoch= N, batch_size=50, verbose=1)
# Predications and Score
y_pred = model.predict(X_test_STD)
score = model.evaluate(X_test_STD, y_test)
Epoch 1/1000 537/537 [==============================] - 0s 164us/step - loss: 1.7979 - accuracy: 0.6499 - mean_pred: 0.0065 Epoch 2/1000 537/537 [==============================] - 0s 19us/step - loss: 1.4491 - accuracy: 0.6499 - mean_pred: 0.0141 Epoch 3/1000 537/537 [==============================] - 0s 19us/step - loss: 1.2474 - accuracy: 0.6499 - mean_pred: 0.0247 Epoch 4/1000 537/537 [==============================] - 0s 17us/step - loss: 1.0799 - accuracy: 0.6499 - mean_pred: 0.0393 Epoch 5/1000 537/537 [==============================] - 0s 17us/step - loss: 0.9383 - accuracy: 0.6499 - mean_pred: 0.0601 Epoch 6/1000 537/537 [==============================] - 0s 17us/step - loss: 0.8212 - accuracy: 0.6499 - mean_pred: 0.0869 Epoch 7/1000 537/537 [==============================] - 0s 15us/step - loss: 0.7214 - accuracy: 0.6499 - mean_pred: 0.1187 Epoch 8/1000 537/537 [==============================] - 0s 19us/step - loss: 0.6362 - accuracy: 0.6555 - mean_pred: 0.1584 Epoch 9/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5691 - accuracy: 0.6946 - mean_pred: 0.2060 Epoch 10/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5249 - accuracy: 0.7207 - mean_pred: 0.2544 Epoch 11/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5445 - accuracy: 0.7467 - mean_pred: 0.3015 Epoch 12/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5287 - accuracy: 0.7635 - mean_pred: 0.3381 Epoch 13/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5251 - accuracy: 0.7561 - mean_pred: 0.3371 Epoch 14/1000 537/537 [==============================] - 0s 15us/step - loss: 0.5177 - accuracy: 0.7561 - mean_pred: 0.3347 Epoch 15/1000 537/537 [==============================] - 0s 15us/step - loss: 0.5151 - accuracy: 0.7598 - mean_pred: 0.3442 Epoch 16/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5334 - accuracy: 0.7635 - mean_pred: 0.3593 Epoch 17/1000 537/537 [==============================] - 0s 19us/step - loss: 0.5301 - accuracy: 0.7635 - mean_pred: 0.3497 Epoch 18/1000 537/537 [==============================] - 0s 15us/step - loss: 0.5256 - accuracy: 0.7672 - mean_pred: 0.3496 Epoch 19/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5225 - accuracy: 0.7784 - mean_pred: 0.3518 Epoch 20/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5205 - accuracy: 0.7821 - mean_pred: 0.3464 Epoch 21/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5187 - accuracy: 0.7784 - mean_pred: 0.3414 Epoch 22/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5154 - accuracy: 0.7765 - mean_pred: 0.3416 Epoch 23/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5181 - accuracy: 0.7765 - mean_pred: 0.3395 Epoch 24/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4991 - accuracy: 0.7803 - mean_pred: 0.3240 Epoch 25/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4947 - accuracy: 0.7709 - mean_pred: 0.3022 Epoch 26/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4925 - accuracy: 0.7747 - mean_pred: 0.3104 Epoch 27/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4904 - accuracy: 0.7784 - mean_pred: 0.3194 Epoch 28/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4884 - accuracy: 0.7803 - mean_pred: 0.3218 Epoch 29/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4874 - accuracy: 0.7877 - mean_pred: 0.3442 Epoch 30/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4893 - accuracy: 0.7765 - mean_pred: 0.3267 Epoch 31/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4920 - accuracy: 0.7803 - mean_pred: 0.3303 Epoch 32/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4864 - accuracy: 0.7821 - mean_pred: 0.3235 Epoch 33/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4858 - accuracy: 0.7840 - mean_pred: 0.3336 Epoch 34/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4866 - accuracy: 0.7821 - mean_pred: 0.3265 Epoch 35/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4865 - accuracy: 0.7877 - mean_pred: 0.3279 Epoch 36/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4854 - accuracy: 0.7821 - mean_pred: 0.3302 Epoch 37/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4834 - accuracy: 0.7858 - mean_pred: 0.3375 Epoch 38/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4816 - accuracy: 0.7821 - mean_pred: 0.3313 Epoch 39/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5027 - accuracy: 0.7858 - mean_pred: 0.3419 Epoch 40/1000 537/537 [==============================] - 0s 17us/step - loss: 0.5034 - accuracy: 0.7858 - mean_pred: 0.3464 Epoch 41/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4996 - accuracy: 0.7858 - mean_pred: 0.3456 Epoch 42/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4998 - accuracy: 0.7858 - mean_pred: 0.3367 Epoch 43/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4990 - accuracy: 0.7896 - mean_pred: 0.3452 Epoch 44/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4974 - accuracy: 0.7877 - mean_pred: 0.3495 Epoch 45/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4985 - accuracy: 0.7858 - mean_pred: 0.3474 Epoch 46/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4973 - accuracy: 0.7840 - mean_pred: 0.3393 Epoch 47/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4976 - accuracy: 0.7803 - mean_pred: 0.3560 Epoch 48/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4956 - accuracy: 0.7821 - mean_pred: 0.3395 Epoch 49/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4953 - accuracy: 0.7784 - mean_pred: 0.3393 Epoch 50/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4960 - accuracy: 0.7803 - mean_pred: 0.3498 Epoch 51/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4943 - accuracy: 0.7821 - mean_pred: 0.3570 Epoch 52/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4948 - accuracy: 0.7784 - mean_pred: 0.3376 Epoch 53/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4945 - accuracy: 0.7821 - mean_pred: 0.3501 Epoch 54/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4951 - accuracy: 0.7784 - mean_pred: 0.3491 Epoch 55/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4942 - accuracy: 0.7821 - mean_pred: 0.3323 Epoch 56/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4937 - accuracy: 0.7784 - mean_pred: 0.3519 Epoch 57/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4938 - accuracy: 0.7803 - mean_pred: 0.3470 Epoch 58/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4949 - accuracy: 0.7803 - mean_pred: 0.3480 Epoch 59/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4930 - accuracy: 0.7821 - mean_pred: 0.3506 Epoch 60/1000 537/537 [==============================] - 0s 13us/step - loss: 0.4931 - accuracy: 0.7765 - mean_pred: 0.3467 Epoch 61/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4935 - accuracy: 0.7803 - mean_pred: 0.3483 Epoch 62/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4924 - accuracy: 0.7821 - mean_pred: 0.3469 Epoch 63/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4933 - accuracy: 0.7784 - mean_pred: 0.3385 Epoch 64/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4926 - accuracy: 0.7784 - mean_pred: 0.3466 Epoch 65/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4919 - accuracy: 0.7821 - mean_pred: 0.3497 Epoch 66/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4922 - accuracy: 0.7821 - mean_pred: 0.3446 Epoch 67/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4917 - accuracy: 0.7858 - mean_pred: 0.3550 Epoch 68/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4926 - accuracy: 0.7747 - mean_pred: 0.3462 Epoch 69/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4915 - accuracy: 0.7784 - mean_pred: 0.3449 Epoch 70/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4911 - accuracy: 0.7896 - mean_pred: 0.3551 Epoch 71/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4914 - accuracy: 0.7765 - mean_pred: 0.3450 Epoch 72/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4912 - accuracy: 0.7765 - mean_pred: 0.3351 Epoch 73/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4900 - accuracy: 0.7858 - mean_pred: 0.3654 Epoch 74/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4903 - accuracy: 0.7765 - mean_pred: 0.3331 Epoch 75/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4902 - accuracy: 0.7840 - mean_pred: 0.3532 Epoch 76/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4900 - accuracy: 0.7821 - mean_pred: 0.3427 Epoch 77/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4902 - accuracy: 0.7784 - mean_pred: 0.3483 Epoch 78/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4893 - accuracy: 0.7821 - mean_pred: 0.3469 Epoch 79/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4900 - accuracy: 0.7784 - mean_pred: 0.3484 Epoch 80/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4898 - accuracy: 0.7840 - mean_pred: 0.3469 Epoch 81/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4898 - accuracy: 0.7877 - mean_pred: 0.3589 Epoch 82/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4912 - accuracy: 0.7784 - mean_pred: 0.3495 Epoch 83/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4902 - accuracy: 0.7821 - mean_pred: 0.3414 Epoch 84/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4889 - accuracy: 0.7803 - mean_pred: 0.3474 Epoch 85/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4879 - accuracy: 0.7840 - mean_pred: 0.3502 Epoch 86/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4881 - accuracy: 0.7765 - mean_pred: 0.3446 Epoch 87/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4891 - accuracy: 0.7765 - mean_pred: 0.3469 Epoch 88/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4887 - accuracy: 0.7784 - mean_pred: 0.3490 Epoch 89/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4887 - accuracy: 0.7821 - mean_pred: 0.3548 Epoch 90/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4902 - accuracy: 0.7784 - mean_pred: 0.3521 Epoch 91/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4877 - accuracy: 0.7821 - mean_pred: 0.3535 Epoch 92/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4878 - accuracy: 0.7803 - mean_pred: 0.3477 Epoch 93/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4882 - accuracy: 0.7784 - mean_pred: 0.3484 Epoch 94/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4892 - accuracy: 0.7747 - mean_pred: 0.3592 Epoch 95/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4868 - accuracy: 0.7803 - mean_pred: 0.3417 Epoch 96/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4874 - accuracy: 0.7821 - mean_pred: 0.3535 Epoch 97/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4881 - accuracy: 0.7803 - mean_pred: 0.3525 Epoch 98/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4874 - accuracy: 0.7803 - mean_pred: 0.3515 Epoch 99/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4870 - accuracy: 0.7821 - mean_pred: 0.3353 Epoch 100/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4877 - accuracy: 0.7821 - mean_pred: 0.3504 Epoch 101/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4867 - accuracy: 0.7803 - mean_pred: 0.3474 Epoch 102/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4864 - accuracy: 0.7803 - mean_pred: 0.3484 Epoch 103/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4866 - accuracy: 0.7821 - mean_pred: 0.3569 Epoch 104/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4856 - accuracy: 0.7858 - mean_pred: 0.3429 Epoch 105/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4879 - accuracy: 0.7784 - mean_pred: 0.3510 Epoch 106/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4860 - accuracy: 0.7765 - mean_pred: 0.3453 Epoch 107/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4853 - accuracy: 0.7803 - mean_pred: 0.3464 Epoch 108/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4862 - accuracy: 0.7784 - mean_pred: 0.3462 Epoch 109/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4857 - accuracy: 0.7784 - mean_pred: 0.3552 Epoch 110/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4850 - accuracy: 0.7784 - mean_pred: 0.3444 Epoch 111/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4851 - accuracy: 0.7840 - mean_pred: 0.3566 Epoch 112/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4843 - accuracy: 0.7821 - mean_pred: 0.3496 Epoch 113/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4853 - accuracy: 0.7765 - mean_pred: 0.3526 Epoch 114/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4850 - accuracy: 0.7840 - mean_pred: 0.3553 Epoch 115/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4848 - accuracy: 0.7821 - mean_pred: 0.3593 Epoch 116/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4836 - accuracy: 0.7803 - mean_pred: 0.3484 Epoch 117/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4854 - accuracy: 0.7821 - mean_pred: 0.3580 Epoch 118/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4846 - accuracy: 0.7784 - mean_pred: 0.3456 Epoch 119/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4835 - accuracy: 0.7803 - mean_pred: 0.3438 Epoch 120/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4843 - accuracy: 0.7858 - mean_pred: 0.3615 Epoch 121/1000 537/537 [==============================] - 0s 13us/step - loss: 0.4838 - accuracy: 0.7803 - mean_pred: 0.3477 Epoch 122/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4837 - accuracy: 0.7803 - mean_pred: 0.3509 Epoch 123/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4827 - accuracy: 0.7821 - mean_pred: 0.3454 Epoch 124/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4837 - accuracy: 0.7858 - mean_pred: 0.3619 Epoch 125/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4844 - accuracy: 0.7821 - mean_pred: 0.3505 Epoch 126/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4842 - accuracy: 0.7803 - mean_pred: 0.3529 Epoch 127/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4816 - accuracy: 0.7840 - mean_pred: 0.3460 Epoch 128/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4832 - accuracy: 0.7840 - mean_pred: 0.3585 Epoch 129/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4829 - accuracy: 0.7821 - mean_pred: 0.3520 Epoch 130/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4821 - accuracy: 0.7803 - mean_pred: 0.3473 Epoch 131/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4818 - accuracy: 0.7896 - mean_pred: 0.3486 Epoch 132/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4831 - accuracy: 0.7821 - mean_pred: 0.3510 Epoch 133/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4833 - accuracy: 0.7784 - mean_pred: 0.3559 Epoch 134/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4809 - accuracy: 0.7821 - mean_pred: 0.3495 Epoch 135/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4814 - accuracy: 0.7784 - mean_pred: 0.3505 Epoch 136/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4825 - accuracy: 0.7858 - mean_pred: 0.3543 Epoch 137/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4819 - accuracy: 0.7858 - mean_pred: 0.3525 Epoch 138/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4806 - accuracy: 0.7803 - mean_pred: 0.3434 Epoch 139/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4816 - accuracy: 0.7858 - mean_pred: 0.3571 Epoch 140/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4798 - accuracy: 0.7877 - mean_pred: 0.3429 Epoch 141/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4812 - accuracy: 0.7821 - mean_pred: 0.3534 Epoch 142/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4804 - accuracy: 0.7858 - mean_pred: 0.3602 Epoch 143/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4803 - accuracy: 0.7877 - mean_pred: 0.3554 Epoch 144/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4793 - accuracy: 0.7877 - mean_pred: 0.3451 Epoch 145/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4805 - accuracy: 0.7803 - mean_pred: 0.3588 Epoch 146/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4806 - accuracy: 0.7858 - mean_pred: 0.3498 Epoch 147/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4799 - accuracy: 0.7858 - mean_pred: 0.3540 Epoch 148/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4795 - accuracy: 0.7858 - mean_pred: 0.3597 Epoch 149/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4807 - accuracy: 0.7803 - mean_pred: 0.3418 Epoch 150/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4790 - accuracy: 0.7877 - mean_pred: 0.3557 Epoch 151/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4786 - accuracy: 0.7840 - mean_pred: 0.3544 Epoch 152/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4805 - accuracy: 0.7821 - mean_pred: 0.3494 Epoch 153/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4801 - accuracy: 0.7858 - mean_pred: 0.3494 Epoch 154/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4791 - accuracy: 0.7858 - mean_pred: 0.3569 Epoch 155/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4786 - accuracy: 0.7765 - mean_pred: 0.3477 Epoch 156/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4796 - accuracy: 0.7877 - mean_pred: 0.3579 Epoch 157/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4794 - accuracy: 0.7840 - mean_pred: 0.3494 Epoch 158/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4785 - accuracy: 0.7840 - mean_pred: 0.3509 Epoch 159/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4799 - accuracy: 0.7877 - mean_pred: 0.3546 Epoch 160/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4779 - accuracy: 0.7896 - mean_pred: 0.3576 Epoch 161/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4789 - accuracy: 0.7914 - mean_pred: 0.3551 Epoch 162/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4778 - accuracy: 0.7858 - mean_pred: 0.3562 Epoch 163/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4773 - accuracy: 0.7840 - mean_pred: 0.3580 Epoch 164/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4802 - accuracy: 0.7877 - mean_pred: 0.3613 Epoch 165/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4775 - accuracy: 0.7858 - mean_pred: 0.3533 Epoch 166/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4771 - accuracy: 0.7840 - mean_pred: 0.3524 Epoch 167/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4775 - accuracy: 0.7877 - mean_pred: 0.3502 Epoch 168/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4771 - accuracy: 0.7896 - mean_pred: 0.3541 Epoch 169/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4759 - accuracy: 0.7896 - mean_pred: 0.3644 Epoch 170/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4768 - accuracy: 0.7858 - mean_pred: 0.3473 Epoch 171/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4772 - accuracy: 0.7821 - mean_pred: 0.3446 Epoch 172/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4760 - accuracy: 0.7896 - mean_pred: 0.3601 Epoch 173/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4755 - accuracy: 0.7858 - mean_pred: 0.3539 Epoch 174/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4762 - accuracy: 0.7914 - mean_pred: 0.3600 Epoch 175/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4747 - accuracy: 0.7914 - mean_pred: 0.3620 Epoch 176/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4750 - accuracy: 0.7858 - mean_pred: 0.3418 Epoch 177/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4759 - accuracy: 0.7877 - mean_pred: 0.3557 Epoch 178/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4748 - accuracy: 0.7914 - mean_pred: 0.3548 Epoch 179/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4761 - accuracy: 0.7914 - mean_pred: 0.3548 Epoch 180/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4741 - accuracy: 0.7933 - mean_pred: 0.3540 Epoch 181/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4768 - accuracy: 0.7877 - mean_pred: 0.3568 Epoch 182/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4738 - accuracy: 0.7896 - mean_pred: 0.3555 Epoch 183/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4739 - accuracy: 0.7896 - mean_pred: 0.3510 Epoch 184/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4745 - accuracy: 0.7858 - mean_pred: 0.3589 Epoch 185/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4740 - accuracy: 0.7877 - mean_pred: 0.3530 Epoch 186/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4737 - accuracy: 0.7970 - mean_pred: 0.3545 Epoch 187/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4739 - accuracy: 0.7914 - mean_pred: 0.3593 Epoch 188/1000 537/537 [==============================] - 0s 13us/step - loss: 0.4734 - accuracy: 0.7914 - mean_pred: 0.3573 Epoch 189/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4733 - accuracy: 0.7896 - mean_pred: 0.3549 Epoch 190/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4736 - accuracy: 0.7933 - mean_pred: 0.3590 Epoch 191/1000 537/537 [==============================] - 0s 13us/step - loss: 0.4732 - accuracy: 0.7933 - mean_pred: 0.3556 Epoch 192/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4736 - accuracy: 0.7933 - mean_pred: 0.3515 Epoch 193/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4736 - accuracy: 0.7877 - mean_pred: 0.3528 Epoch 194/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4718 - accuracy: 0.7933 - mean_pred: 0.3617 Epoch 195/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4724 - accuracy: 0.7933 - mean_pred: 0.3609 Epoch 196/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4712 - accuracy: 0.7970 - mean_pred: 0.3480 Epoch 197/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4737 - accuracy: 0.7933 - mean_pred: 0.3596 Epoch 198/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4724 - accuracy: 0.7952 - mean_pred: 0.3517 Epoch 199/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4716 - accuracy: 0.7933 - mean_pred: 0.3630 Epoch 200/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4714 - accuracy: 0.7914 - mean_pred: 0.3499 Epoch 201/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4720 - accuracy: 0.7989 - mean_pred: 0.3575 Epoch 202/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4702 - accuracy: 0.8026 - mean_pred: 0.3572 Epoch 203/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4711 - accuracy: 0.7970 - mean_pred: 0.3523 Epoch 204/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4707 - accuracy: 0.7989 - mean_pred: 0.3604 Epoch 205/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4716 - accuracy: 0.7952 - mean_pred: 0.3544 Epoch 206/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4702 - accuracy: 0.7970 - mean_pred: 0.3564 Epoch 207/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4704 - accuracy: 0.7952 - mean_pred: 0.3595 Epoch 208/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4700 - accuracy: 0.7970 - mean_pred: 0.3539 Epoch 209/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4698 - accuracy: 0.7970 - mean_pred: 0.3592 Epoch 210/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4699 - accuracy: 0.7989 - mean_pred: 0.3521 Epoch 211/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4705 - accuracy: 0.7970 - mean_pred: 0.3609 Epoch 212/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4689 - accuracy: 0.8026 - mean_pred: 0.3549 Epoch 213/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4685 - accuracy: 0.7989 - mean_pred: 0.3577 Epoch 214/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4696 - accuracy: 0.7933 - mean_pred: 0.3598 Epoch 215/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4688 - accuracy: 0.7933 - mean_pred: 0.3560 Epoch 216/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4690 - accuracy: 0.7989 - mean_pred: 0.3558 Epoch 217/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4689 - accuracy: 0.8007 - mean_pred: 0.3592 Epoch 218/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4691 - accuracy: 0.7896 - mean_pred: 0.3568 Epoch 219/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4677 - accuracy: 0.7952 - mean_pred: 0.3503 Epoch 220/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4679 - accuracy: 0.7933 - mean_pred: 0.3649 Epoch 221/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4678 - accuracy: 0.7952 - mean_pred: 0.3500 Epoch 222/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4670 - accuracy: 0.7933 - mean_pred: 0.3630 Epoch 223/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4684 - accuracy: 0.7970 - mean_pred: 0.3514 Epoch 224/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4670 - accuracy: 0.7877 - mean_pred: 0.3637 Epoch 225/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4672 - accuracy: 0.7989 - mean_pred: 0.3617 Epoch 226/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4671 - accuracy: 0.7970 - mean_pred: 0.3563 Epoch 227/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4668 - accuracy: 0.8045 - mean_pred: 0.3604 Epoch 228/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4673 - accuracy: 0.7970 - mean_pred: 0.3507 Epoch 229/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4667 - accuracy: 0.7952 - mean_pred: 0.3589 Epoch 230/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4653 - accuracy: 0.7914 - mean_pred: 0.3641 Epoch 231/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4663 - accuracy: 0.8045 - mean_pred: 0.3469 Epoch 232/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4662 - accuracy: 0.7877 - mean_pred: 0.3721 Epoch 233/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4662 - accuracy: 0.8007 - mean_pred: 0.3522 Epoch 234/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4661 - accuracy: 0.8026 - mean_pred: 0.3560 Epoch 235/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4659 - accuracy: 0.7970 - mean_pred: 0.3556 Epoch 236/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4655 - accuracy: 0.7933 - mean_pred: 0.3549 Epoch 237/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4661 - accuracy: 0.8007 - mean_pred: 0.3633 Epoch 238/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4653 - accuracy: 0.7952 - mean_pred: 0.3530 Epoch 239/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4657 - accuracy: 0.8007 - mean_pred: 0.3520 Epoch 240/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4643 - accuracy: 0.7933 - mean_pred: 0.3635 Epoch 241/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4652 - accuracy: 0.7989 - mean_pred: 0.3472 Epoch 242/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4653 - accuracy: 0.7952 - mean_pred: 0.3667 Epoch 243/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4667 - accuracy: 0.7933 - mean_pred: 0.3625 Epoch 244/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4641 - accuracy: 0.7989 - mean_pred: 0.3563 Epoch 245/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4640 - accuracy: 0.7970 - mean_pred: 0.3565 Epoch 246/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4640 - accuracy: 0.7970 - mean_pred: 0.3497 Epoch 247/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4639 - accuracy: 0.7970 - mean_pred: 0.3587 Epoch 248/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4644 - accuracy: 0.7952 - mean_pred: 0.3604 Epoch 249/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4636 - accuracy: 0.8026 - mean_pred: 0.3540 Epoch 250/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4625 - accuracy: 0.7989 - mean_pred: 0.3627 Epoch 251/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4634 - accuracy: 0.7989 - mean_pred: 0.3522 Epoch 252/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4630 - accuracy: 0.8045 - mean_pred: 0.3538 Epoch 253/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4649 - accuracy: 0.8007 - mean_pred: 0.3622 Epoch 254/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4616 - accuracy: 0.8026 - mean_pred: 0.3586 Epoch 255/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4653 - accuracy: 0.7970 - mean_pred: 0.3586 Epoch 256/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4610 - accuracy: 0.8045 - mean_pred: 0.3577 Epoch 257/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4623 - accuracy: 0.7989 - mean_pred: 0.3538 Epoch 258/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4638 - accuracy: 0.7914 - mean_pred: 0.3683 Epoch 259/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4620 - accuracy: 0.7989 - mean_pred: 0.3489 Epoch 260/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4626 - accuracy: 0.8026 - mean_pred: 0.3545 Epoch 261/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4613 - accuracy: 0.8045 - mean_pred: 0.3548 Epoch 262/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4618 - accuracy: 0.7970 - mean_pred: 0.3657 Epoch 263/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4627 - accuracy: 0.8026 - mean_pred: 0.3516 Epoch 264/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4605 - accuracy: 0.8007 - mean_pred: 0.3630 Epoch 265/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4610 - accuracy: 0.8026 - mean_pred: 0.3587 Epoch 266/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4609 - accuracy: 0.8007 - mean_pred: 0.3519 Epoch 267/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4594 - accuracy: 0.8007 - mean_pred: 0.3638 Epoch 268/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4625 - accuracy: 0.8026 - mean_pred: 0.3529 Epoch 269/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4421 - accuracy: 0.8045 - mean_pred: 0.3587 Epoch 270/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4221 - accuracy: 0.8026 - mean_pred: 0.3173 Epoch 271/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4181 - accuracy: 0.8045 - mean_pred: 0.3159 Epoch 272/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4163 - accuracy: 0.8007 - mean_pred: 0.3300 Epoch 273/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4170 - accuracy: 0.7952 - mean_pred: 0.3381 Epoch 274/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4154 - accuracy: 0.7989 - mean_pred: 0.3378 Epoch 275/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4179 - accuracy: 0.8026 - mean_pred: 0.3443 Epoch 276/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4164 - accuracy: 0.7989 - mean_pred: 0.3381 Epoch 277/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4153 - accuracy: 0.7952 - mean_pred: 0.3425 Epoch 278/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4152 - accuracy: 0.7933 - mean_pred: 0.3421 Epoch 279/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4144 - accuracy: 0.8007 - mean_pred: 0.3400 Epoch 280/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4155 - accuracy: 0.8007 - mean_pred: 0.3415 Epoch 281/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4147 - accuracy: 0.8063 - mean_pred: 0.3537 Epoch 282/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4154 - accuracy: 0.7989 - mean_pred: 0.3435 Epoch 283/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4132 - accuracy: 0.8007 - mean_pred: 0.3436 Epoch 284/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4138 - accuracy: 0.8045 - mean_pred: 0.3499 Epoch 285/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4138 - accuracy: 0.8007 - mean_pred: 0.3407 Epoch 286/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4148 - accuracy: 0.7933 - mean_pred: 0.3358 Epoch 287/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4135 - accuracy: 0.8063 - mean_pred: 0.3528 Epoch 288/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4130 - accuracy: 0.8026 - mean_pred: 0.3462 Epoch 289/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4136 - accuracy: 0.7952 - mean_pred: 0.3485 Epoch 290/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4136 - accuracy: 0.8082 - mean_pred: 0.3492 Epoch 291/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4124 - accuracy: 0.8063 - mean_pred: 0.3506 Epoch 292/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4126 - accuracy: 0.8045 - mean_pred: 0.3428 Epoch 293/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4145 - accuracy: 0.8026 - mean_pred: 0.3541 Epoch 294/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4117 - accuracy: 0.8026 - mean_pred: 0.3440 Epoch 295/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4121 - accuracy: 0.8063 - mean_pred: 0.3456 Epoch 296/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4112 - accuracy: 0.8082 - mean_pred: 0.3489 Epoch 297/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4136 - accuracy: 0.8045 - mean_pred: 0.3568 Epoch 298/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4118 - accuracy: 0.8082 - mean_pred: 0.3507 Epoch 299/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4111 - accuracy: 0.7989 - mean_pred: 0.3540 Epoch 300/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4116 - accuracy: 0.8063 - mean_pred: 0.3544 Epoch 301/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4124 - accuracy: 0.7914 - mean_pred: 0.3499 Epoch 302/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4112 - accuracy: 0.8026 - mean_pred: 0.3405 Epoch 303/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4101 - accuracy: 0.8045 - mean_pred: 0.3417 Epoch 304/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4112 - accuracy: 0.8026 - mean_pred: 0.3578 Epoch 305/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4140 - accuracy: 0.8007 - mean_pred: 0.3515 Epoch 306/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4100 - accuracy: 0.8045 - mean_pred: 0.3506 Epoch 307/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4109 - accuracy: 0.8007 - mean_pred: 0.3524 Epoch 308/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4102 - accuracy: 0.8045 - mean_pred: 0.3523 Epoch 309/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4093 - accuracy: 0.8082 - mean_pred: 0.3534 Epoch 310/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4104 - accuracy: 0.8007 - mean_pred: 0.3495 Epoch 311/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4123 - accuracy: 0.8045 - mean_pred: 0.3470 Epoch 312/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4103 - accuracy: 0.8045 - mean_pred: 0.3493 Epoch 313/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4100 - accuracy: 0.8026 - mean_pred: 0.3576 Epoch 314/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4101 - accuracy: 0.8063 - mean_pred: 0.3497 Epoch 315/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4097 - accuracy: 0.8082 - mean_pred: 0.3456 Epoch 316/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4102 - accuracy: 0.8007 - mean_pred: 0.3520 Epoch 317/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4106 - accuracy: 0.7989 - mean_pred: 0.3514 Epoch 318/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4086 - accuracy: 0.8007 - mean_pred: 0.3534 Epoch 319/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4103 - accuracy: 0.8026 - mean_pred: 0.3500 Epoch 320/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4126 - accuracy: 0.8063 - mean_pred: 0.3619 Epoch 321/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4080 - accuracy: 0.8063 - mean_pred: 0.3399 Epoch 322/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4090 - accuracy: 0.8007 - mean_pred: 0.3620 Epoch 323/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4094 - accuracy: 0.8063 - mean_pred: 0.3479 Epoch 324/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4087 - accuracy: 0.8101 - mean_pred: 0.3521 Epoch 325/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4093 - accuracy: 0.7970 - mean_pred: 0.3603 Epoch 326/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4089 - accuracy: 0.8082 - mean_pred: 0.3466 Epoch 327/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4110 - accuracy: 0.8026 - mean_pred: 0.3657 Epoch 328/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4090 - accuracy: 0.8063 - mean_pred: 0.3471 Epoch 329/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4095 - accuracy: 0.8026 - mean_pred: 0.3559 Epoch 330/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4087 - accuracy: 0.8007 - mean_pred: 0.3521 Epoch 331/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4091 - accuracy: 0.7970 - mean_pred: 0.3483 Epoch 332/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4071 - accuracy: 0.8082 - mean_pred: 0.3615 Epoch 333/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4105 - accuracy: 0.8138 - mean_pred: 0.3459 Epoch 334/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4099 - accuracy: 0.8007 - mean_pred: 0.3418 Epoch 335/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4090 - accuracy: 0.8045 - mean_pred: 0.3470 Epoch 336/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4090 - accuracy: 0.8119 - mean_pred: 0.3592 Epoch 337/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4076 - accuracy: 0.8045 - mean_pred: 0.3529 Epoch 338/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4090 - accuracy: 0.8063 - mean_pred: 0.3501 Epoch 339/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4103 - accuracy: 0.8119 - mean_pred: 0.3590 Epoch 340/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4071 - accuracy: 0.8026 - mean_pred: 0.3477 Epoch 341/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4089 - accuracy: 0.8026 - mean_pred: 0.3524 Epoch 342/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4083 - accuracy: 0.8063 - mean_pred: 0.3547 Epoch 343/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4076 - accuracy: 0.8063 - mean_pred: 0.3508 Epoch 344/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4079 - accuracy: 0.8045 - mean_pred: 0.3519 Epoch 345/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4100 - accuracy: 0.8045 - mean_pred: 0.3518 Epoch 346/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4062 - accuracy: 0.8045 - mean_pred: 0.3503 Epoch 347/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4080 - accuracy: 0.8026 - mean_pred: 0.3577 Epoch 348/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4077 - accuracy: 0.8045 - mean_pred: 0.3484 Epoch 349/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4079 - accuracy: 0.8063 - mean_pred: 0.3452 Epoch 350/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4086 - accuracy: 0.8007 - mean_pred: 0.3546 Epoch 351/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4070 - accuracy: 0.8063 - mean_pred: 0.3506 Epoch 352/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4079 - accuracy: 0.8063 - mean_pred: 0.3477 Epoch 353/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4077 - accuracy: 0.7989 - mean_pred: 0.3578 Epoch 354/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4072 - accuracy: 0.8026 - mean_pred: 0.3521 Epoch 355/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4064 - accuracy: 0.8101 - mean_pred: 0.3546 Epoch 356/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4086 - accuracy: 0.8101 - mean_pred: 0.3557 Epoch 357/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4073 - accuracy: 0.8138 - mean_pred: 0.3554 Epoch 358/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4054 - accuracy: 0.8063 - mean_pred: 0.3603 Epoch 359/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4046 - accuracy: 0.8007 - mean_pred: 0.3400 Epoch 360/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4077 - accuracy: 0.8045 - mean_pred: 0.3688 Epoch 361/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4072 - accuracy: 0.8082 - mean_pred: 0.3637 Epoch 362/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4063 - accuracy: 0.8082 - mean_pred: 0.3479 Epoch 363/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4056 - accuracy: 0.8045 - mean_pred: 0.3500 Epoch 364/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4071 - accuracy: 0.8063 - mean_pred: 0.3550 Epoch 365/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4096 - accuracy: 0.8063 - mean_pred: 0.3521 Epoch 366/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4058 - accuracy: 0.8007 - mean_pred: 0.3558 Epoch 367/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4071 - accuracy: 0.8007 - mean_pred: 0.3529 Epoch 368/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4054 - accuracy: 0.8119 - mean_pred: 0.3462 Epoch 369/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4067 - accuracy: 0.8063 - mean_pred: 0.3597 Epoch 370/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4057 - accuracy: 0.8101 - mean_pred: 0.3581 Epoch 371/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4067 - accuracy: 0.8026 - mean_pred: 0.3465 Epoch 372/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4063 - accuracy: 0.8026 - mean_pred: 0.3563 Epoch 373/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4061 - accuracy: 0.8101 - mean_pred: 0.3516 Epoch 374/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4080 - accuracy: 0.8082 - mean_pred: 0.3454 Epoch 375/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4061 - accuracy: 0.8101 - mean_pred: 0.3567 Epoch 376/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4062 - accuracy: 0.8045 - mean_pred: 0.3595 Epoch 377/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4066 - accuracy: 0.8119 - mean_pred: 0.3496 Epoch 378/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4057 - accuracy: 0.8007 - mean_pred: 0.3537 Epoch 379/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4064 - accuracy: 0.8045 - mean_pred: 0.3531 Epoch 380/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4050 - accuracy: 0.8063 - mean_pred: 0.3550 Epoch 381/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4055 - accuracy: 0.8101 - mean_pred: 0.3600 Epoch 382/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4059 - accuracy: 0.8119 - mean_pred: 0.3507 Epoch 383/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4068 - accuracy: 0.8101 - mean_pred: 0.3512 Epoch 384/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4040 - accuracy: 0.8082 - mean_pred: 0.3629 Epoch 385/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4066 - accuracy: 0.8063 - mean_pred: 0.3487 Epoch 386/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4055 - accuracy: 0.8082 - mean_pred: 0.3488 Epoch 387/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4054 - accuracy: 0.8045 - mean_pred: 0.3591 Epoch 388/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4052 - accuracy: 0.8101 - mean_pred: 0.3522 Epoch 389/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4053 - accuracy: 0.8045 - mean_pred: 0.3549 Epoch 390/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4049 - accuracy: 0.8082 - mean_pred: 0.3598 Epoch 391/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4044 - accuracy: 0.8101 - mean_pred: 0.3480 Epoch 392/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4053 - accuracy: 0.8082 - mean_pred: 0.3607 Epoch 393/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4022 - accuracy: 0.8045 - mean_pred: 0.3474 Epoch 394/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4059 - accuracy: 0.8045 - mean_pred: 0.3658 Epoch 395/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4053 - accuracy: 0.8026 - mean_pred: 0.3547 Epoch 396/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4049 - accuracy: 0.8007 - mean_pred: 0.3490 Epoch 397/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4059 - accuracy: 0.8045 - mean_pred: 0.3565 Epoch 398/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4050 - accuracy: 0.7989 - mean_pred: 0.3529 Epoch 399/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4048 - accuracy: 0.8063 - mean_pred: 0.3650 Epoch 400/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4052 - accuracy: 0.8063 - mean_pred: 0.3537 Epoch 401/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4042 - accuracy: 0.8063 - mean_pred: 0.3514 Epoch 402/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4040 - accuracy: 0.8007 - mean_pred: 0.3554 Epoch 403/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4052 - accuracy: 0.8082 - mean_pred: 0.3566 Epoch 404/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4042 - accuracy: 0.8082 - mean_pred: 0.3452 Epoch 405/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4037 - accuracy: 0.8026 - mean_pred: 0.3585 Epoch 406/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4042 - accuracy: 0.8045 - mean_pred: 0.3598 Epoch 407/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4054 - accuracy: 0.8007 - mean_pred: 0.3459 Epoch 408/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4039 - accuracy: 0.8082 - mean_pred: 0.3624 Epoch 409/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4031 - accuracy: 0.8026 - mean_pred: 0.3494 Epoch 410/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4045 - accuracy: 0.8101 - mean_pred: 0.3616 Epoch 411/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4050 - accuracy: 0.8063 - mean_pred: 0.3520 Epoch 412/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4029 - accuracy: 0.8045 - mean_pred: 0.3510 Epoch 413/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4046 - accuracy: 0.8063 - mean_pred: 0.3597 Epoch 414/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4029 - accuracy: 0.8063 - mean_pred: 0.3551 Epoch 415/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4022 - accuracy: 0.8045 - mean_pred: 0.3648 Epoch 416/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4042 - accuracy: 0.8101 - mean_pred: 0.3446 Epoch 417/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4044 - accuracy: 0.8007 - mean_pred: 0.3554 Epoch 418/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4031 - accuracy: 0.8101 - mean_pred: 0.3560 Epoch 419/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4041 - accuracy: 0.8101 - mean_pred: 0.3586 Epoch 420/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4035 - accuracy: 0.8101 - mean_pred: 0.3568 Epoch 421/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4023 - accuracy: 0.8101 - mean_pred: 0.3525 Epoch 422/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4057 - accuracy: 0.8045 - mean_pred: 0.3633 Epoch 423/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4034 - accuracy: 0.8119 - mean_pred: 0.3555 Epoch 424/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4028 - accuracy: 0.8045 - mean_pred: 0.3551 Epoch 425/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4030 - accuracy: 0.8063 - mean_pred: 0.3556 Epoch 426/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4046 - accuracy: 0.8082 - mean_pred: 0.3552 Epoch 427/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4039 - accuracy: 0.8101 - mean_pred: 0.3568 Epoch 428/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4034 - accuracy: 0.8063 - mean_pred: 0.3554 Epoch 429/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4031 - accuracy: 0.8063 - mean_pred: 0.3550 Epoch 430/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4030 - accuracy: 0.8007 - mean_pred: 0.3550 Epoch 431/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4048 - accuracy: 0.8026 - mean_pred: 0.3550 Epoch 432/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4012 - accuracy: 0.8082 - mean_pred: 0.3546 Epoch 433/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4022 - accuracy: 0.8007 - mean_pred: 0.3589 Epoch 434/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4027 - accuracy: 0.8007 - mean_pred: 0.3582 Epoch 435/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4030 - accuracy: 0.8063 - mean_pred: 0.3596 Epoch 436/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4026 - accuracy: 0.8063 - mean_pred: 0.3444 Epoch 437/1000 537/537 [==============================] - ETA: 0s - loss: 0.6637 - accuracy: 0.8200 - mean_pred: 0.31 - 0s 15us/step - loss: 0.4030 - accuracy: 0.8007 - mean_pred: 0.3570 Epoch 438/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4032 - accuracy: 0.8063 - mean_pred: 0.3557 Epoch 439/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4006 - accuracy: 0.8101 - mean_pred: 0.3665 Epoch 440/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4024 - accuracy: 0.8119 - mean_pred: 0.3536 Epoch 441/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4031 - accuracy: 0.8082 - mean_pred: 0.3649 Epoch 442/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4016 - accuracy: 0.8063 - mean_pred: 0.3528 Epoch 443/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4012 - accuracy: 0.8007 - mean_pred: 0.3471 Epoch 444/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4025 - accuracy: 0.8045 - mean_pred: 0.3635 Epoch 445/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4032 - accuracy: 0.8045 - mean_pred: 0.3598 Epoch 446/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4035 - accuracy: 0.8026 - mean_pred: 0.3614 Epoch 447/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4024 - accuracy: 0.8045 - mean_pred: 0.3514 Epoch 448/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4017 - accuracy: 0.8026 - mean_pred: 0.3569 Epoch 449/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4018 - accuracy: 0.8045 - mean_pred: 0.3533 Epoch 450/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4013 - accuracy: 0.8101 - mean_pred: 0.3554 Epoch 451/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4021 - accuracy: 0.8138 - mean_pred: 0.3533 Epoch 452/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4029 - accuracy: 0.8045 - mean_pred: 0.3669 Epoch 453/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4016 - accuracy: 0.8082 - mean_pred: 0.3536 Epoch 454/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4012 - accuracy: 0.8101 - mean_pred: 0.3574 Epoch 455/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4018 - accuracy: 0.8101 - mean_pred: 0.3534 Epoch 456/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4008 - accuracy: 0.8007 - mean_pred: 0.3635 Epoch 457/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4041 - accuracy: 0.8101 - mean_pred: 0.3566 Epoch 458/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3991 - accuracy: 0.7989 - mean_pred: 0.3509 Epoch 459/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3992 - accuracy: 0.8045 - mean_pred: 0.3501 Epoch 460/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3999 - accuracy: 0.8082 - mean_pred: 0.3537 Epoch 461/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3999 - accuracy: 0.8045 - mean_pred: 0.3608 Epoch 462/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4011 - accuracy: 0.8045 - mean_pred: 0.3509 Epoch 463/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4014 - accuracy: 0.8063 - mean_pred: 0.3560 Epoch 464/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4004 - accuracy: 0.8119 - mean_pred: 0.3565 Epoch 465/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4018 - accuracy: 0.8045 - mean_pred: 0.3532 Epoch 466/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4010 - accuracy: 0.8045 - mean_pred: 0.3549 Epoch 467/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4039 - accuracy: 0.8026 - mean_pred: 0.3511 Epoch 468/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3996 - accuracy: 0.8026 - mean_pred: 0.3596 Epoch 469/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3995 - accuracy: 0.8063 - mean_pred: 0.3498 Epoch 470/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4008 - accuracy: 0.7989 - mean_pred: 0.3617 Epoch 471/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4002 - accuracy: 0.8045 - mean_pred: 0.3536 Epoch 472/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4020 - accuracy: 0.7989 - mean_pred: 0.3562 Epoch 473/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4009 - accuracy: 0.8082 - mean_pred: 0.3505 Epoch 474/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3996 - accuracy: 0.8119 - mean_pred: 0.3575 Epoch 475/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4015 - accuracy: 0.8007 - mean_pred: 0.3466 Epoch 476/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4033 - accuracy: 0.8082 - mean_pred: 0.3635 Epoch 477/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3994 - accuracy: 0.8063 - mean_pred: 0.3551 Epoch 478/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4009 - accuracy: 0.8082 - mean_pred: 0.3506 Epoch 479/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3999 - accuracy: 0.8063 - mean_pred: 0.3607 Epoch 480/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4000 - accuracy: 0.8063 - mean_pred: 0.3459 Epoch 481/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4003 - accuracy: 0.8082 - mean_pred: 0.3591 Epoch 482/1000 537/537 [==============================] - 0s 19us/step - loss: 0.4053 - accuracy: 0.8101 - mean_pred: 0.3593 Epoch 483/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3978 - accuracy: 0.8026 - mean_pred: 0.3516 Epoch 484/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3979 - accuracy: 0.8045 - mean_pred: 0.3533 Epoch 485/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3984 - accuracy: 0.8026 - mean_pred: 0.3543 Epoch 486/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3985 - accuracy: 0.8045 - mean_pred: 0.3530 Epoch 487/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4000 - accuracy: 0.8063 - mean_pred: 0.3595 Epoch 488/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3998 - accuracy: 0.8045 - mean_pred: 0.3530 Epoch 489/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4002 - accuracy: 0.8063 - mean_pred: 0.3562 Epoch 490/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3995 - accuracy: 0.8063 - mean_pred: 0.3542 Epoch 491/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4002 - accuracy: 0.8026 - mean_pred: 0.3636 Epoch 492/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4000 - accuracy: 0.8045 - mean_pred: 0.3526 Epoch 493/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4015 - accuracy: 0.8119 - mean_pred: 0.3582 Epoch 494/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3998 - accuracy: 0.8082 - mean_pred: 0.3497 Epoch 495/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3996 - accuracy: 0.8082 - mean_pred: 0.3589 Epoch 496/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3990 - accuracy: 0.7970 - mean_pred: 0.3478 Epoch 497/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3983 - accuracy: 0.8045 - mean_pred: 0.3637 Epoch 498/1000 537/537 [==============================] - 0s 15us/step - loss: 0.4014 - accuracy: 0.8063 - mean_pred: 0.3603 Epoch 499/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3993 - accuracy: 0.8082 - mean_pred: 0.3506 Epoch 500/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3997 - accuracy: 0.8082 - mean_pred: 0.3406 Epoch 501/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3998 - accuracy: 0.8045 - mean_pred: 0.3638 Epoch 502/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3990 - accuracy: 0.8026 - mean_pred: 0.3464 Epoch 503/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3993 - accuracy: 0.8063 - mean_pred: 0.3536 Epoch 504/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3994 - accuracy: 0.8026 - mean_pred: 0.3574 Epoch 505/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4003 - accuracy: 0.8007 - mean_pred: 0.3601 Epoch 506/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3982 - accuracy: 0.8063 - mean_pred: 0.3549 Epoch 507/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3983 - accuracy: 0.8026 - mean_pred: 0.3568 Epoch 508/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3992 - accuracy: 0.8119 - mean_pred: 0.3594 Epoch 509/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3989 - accuracy: 0.8045 - mean_pred: 0.3507 Epoch 510/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4008 - accuracy: 0.8026 - mean_pred: 0.3566 Epoch 511/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3995 - accuracy: 0.8045 - mean_pred: 0.3573 Epoch 512/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3984 - accuracy: 0.8007 - mean_pred: 0.3569 Epoch 513/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3980 - accuracy: 0.8026 - mean_pred: 0.3557 Epoch 514/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3990 - accuracy: 0.7952 - mean_pred: 0.3502 Epoch 515/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3974 - accuracy: 0.8063 - mean_pred: 0.3630 Epoch 516/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3983 - accuracy: 0.7989 - mean_pred: 0.3549 Epoch 517/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3981 - accuracy: 0.8026 - mean_pred: 0.3513 Epoch 518/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3985 - accuracy: 0.8045 - mean_pred: 0.3590 Epoch 519/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3986 - accuracy: 0.8082 - mean_pred: 0.3626 Epoch 520/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3973 - accuracy: 0.8045 - mean_pred: 0.3467 Epoch 521/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3996 - accuracy: 0.8063 - mean_pred: 0.3606 Epoch 522/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3976 - accuracy: 0.8007 - mean_pred: 0.3579 Epoch 523/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3995 - accuracy: 0.8063 - mean_pred: 0.3522 Epoch 524/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3974 - accuracy: 0.8026 - mean_pred: 0.3483 Epoch 525/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3990 - accuracy: 0.8007 - mean_pred: 0.3656 Epoch 526/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3974 - accuracy: 0.8026 - mean_pred: 0.3476 Epoch 527/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3967 - accuracy: 0.8063 - mean_pred: 0.3539 Epoch 528/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3965 - accuracy: 0.8045 - mean_pred: 0.3551 Epoch 529/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3988 - accuracy: 0.8063 - mean_pred: 0.3600 Epoch 530/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3973 - accuracy: 0.8026 - mean_pred: 0.3584 Epoch 531/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3966 - accuracy: 0.8156 - mean_pred: 0.3676 Epoch 532/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3978 - accuracy: 0.8026 - mean_pred: 0.3446 Epoch 533/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3958 - accuracy: 0.8082 - mean_pred: 0.3562 Epoch 534/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3965 - accuracy: 0.8101 - mean_pred: 0.3615 Epoch 535/1000 537/537 [==============================] - 0s 17us/step - loss: 0.4002 - accuracy: 0.7989 - mean_pred: 0.3583 Epoch 536/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3963 - accuracy: 0.8101 - mean_pred: 0.3494 Epoch 537/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3963 - accuracy: 0.8045 - mean_pred: 0.3615 Epoch 538/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3958 - accuracy: 0.8026 - mean_pred: 0.3542 Epoch 539/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3973 - accuracy: 0.8082 - mean_pred: 0.3571 Epoch 540/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3970 - accuracy: 0.8045 - mean_pred: 0.3610 Epoch 541/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3978 - accuracy: 0.8026 - mean_pred: 0.3550 Epoch 542/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3964 - accuracy: 0.7989 - mean_pred: 0.3521 Epoch 543/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3967 - accuracy: 0.7989 - mean_pred: 0.3592 Epoch 544/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3966 - accuracy: 0.8082 - mean_pred: 0.3610 Epoch 545/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3962 - accuracy: 0.8045 - mean_pred: 0.3513 Epoch 546/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3953 - accuracy: 0.8026 - mean_pred: 0.3491 Epoch 547/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3978 - accuracy: 0.8026 - mean_pred: 0.3721 Epoch 548/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3953 - accuracy: 0.8026 - mean_pred: 0.3611 Epoch 549/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3969 - accuracy: 0.8045 - mean_pred: 0.3504 Epoch 550/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3972 - accuracy: 0.7989 - mean_pred: 0.3522 Epoch 551/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3962 - accuracy: 0.8063 - mean_pred: 0.3581 Epoch 552/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3955 - accuracy: 0.8026 - mean_pred: 0.3492 Epoch 553/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3956 - accuracy: 0.8026 - mean_pred: 0.3601 Epoch 554/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3963 - accuracy: 0.7914 - mean_pred: 0.3551 Epoch 555/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3944 - accuracy: 0.8063 - mean_pred: 0.3548 Epoch 556/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3964 - accuracy: 0.8007 - mean_pred: 0.3660 Epoch 557/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3940 - accuracy: 0.8101 - mean_pred: 0.3469 Epoch 558/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3971 - accuracy: 0.8045 - mean_pred: 0.3667 Epoch 559/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3952 - accuracy: 0.7989 - mean_pred: 0.3610 Epoch 560/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3970 - accuracy: 0.8045 - mean_pred: 0.3525 Epoch 561/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3961 - accuracy: 0.8007 - mean_pred: 0.3537 Epoch 562/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3951 - accuracy: 0.7989 - mean_pred: 0.3623 Epoch 563/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3955 - accuracy: 0.7989 - mean_pred: 0.3562 Epoch 564/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3970 - accuracy: 0.8007 - mean_pred: 0.3579 Epoch 565/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3960 - accuracy: 0.7952 - mean_pred: 0.3536 Epoch 566/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3948 - accuracy: 0.8026 - mean_pred: 0.3596 Epoch 567/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3972 - accuracy: 0.8007 - mean_pred: 0.3578 Epoch 568/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3950 - accuracy: 0.7952 - mean_pred: 0.3583 Epoch 569/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3958 - accuracy: 0.8007 - mean_pred: 0.3555 Epoch 570/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3970 - accuracy: 0.8026 - mean_pred: 0.3636 Epoch 571/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3957 - accuracy: 0.8007 - mean_pred: 0.3585 Epoch 572/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3945 - accuracy: 0.8101 - mean_pred: 0.3492 Epoch 573/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3944 - accuracy: 0.8007 - mean_pred: 0.3600 Epoch 574/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3945 - accuracy: 0.8007 - mean_pred: 0.3549 Epoch 575/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3944 - accuracy: 0.7989 - mean_pred: 0.3510 Epoch 576/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3955 - accuracy: 0.8007 - mean_pred: 0.3608 Epoch 577/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3947 - accuracy: 0.7952 - mean_pred: 0.3643 Epoch 578/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3968 - accuracy: 0.7970 - mean_pred: 0.3514 Epoch 579/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3934 - accuracy: 0.7970 - mean_pred: 0.3672 Epoch 580/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3937 - accuracy: 0.8007 - mean_pred: 0.3509 Epoch 581/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3947 - accuracy: 0.8045 - mean_pred: 0.3591 Epoch 582/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3933 - accuracy: 0.8026 - mean_pred: 0.3592 Epoch 583/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3926 - accuracy: 0.7989 - mean_pred: 0.3549 Epoch 584/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3953 - accuracy: 0.7952 - mean_pred: 0.3582 Epoch 585/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3961 - accuracy: 0.7970 - mean_pred: 0.3652 Epoch 586/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3940 - accuracy: 0.7952 - mean_pred: 0.3561 Epoch 587/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3941 - accuracy: 0.7970 - mean_pred: 0.3497 Epoch 588/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3931 - accuracy: 0.7989 - mean_pred: 0.3559 Epoch 589/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3938 - accuracy: 0.7970 - mean_pred: 0.3607 Epoch 590/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3934 - accuracy: 0.8045 - mean_pred: 0.3525 Epoch 591/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3937 - accuracy: 0.7933 - mean_pred: 0.3683 Epoch 592/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3935 - accuracy: 0.7952 - mean_pred: 0.3561 Epoch 593/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3926 - accuracy: 0.7989 - mean_pred: 0.3530 Epoch 594/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3952 - accuracy: 0.7933 - mean_pred: 0.3582 Epoch 595/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3941 - accuracy: 0.7933 - mean_pred: 0.3627 Epoch 596/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3935 - accuracy: 0.7970 - mean_pred: 0.3663 Epoch 597/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3938 - accuracy: 0.7952 - mean_pred: 0.3502 Epoch 598/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3936 - accuracy: 0.7970 - mean_pred: 0.3585 Epoch 599/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3922 - accuracy: 0.7970 - mean_pred: 0.3643 Epoch 600/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3924 - accuracy: 0.7952 - mean_pred: 0.3560 Epoch 601/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3921 - accuracy: 0.7952 - mean_pred: 0.3540 Epoch 602/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3921 - accuracy: 0.7933 - mean_pred: 0.3569 Epoch 603/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3937 - accuracy: 0.7914 - mean_pred: 0.3588 Epoch 604/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3930 - accuracy: 0.7952 - mean_pred: 0.3610 Epoch 605/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3913 - accuracy: 0.7914 - mean_pred: 0.3574 Epoch 606/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3913 - accuracy: 0.8063 - mean_pred: 0.3531 Epoch 607/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3917 - accuracy: 0.7952 - mean_pred: 0.3563 Epoch 608/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3939 - accuracy: 0.7952 - mean_pred: 0.3590 Epoch 609/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3906 - accuracy: 0.7933 - mean_pred: 0.3570 Epoch 610/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3909 - accuracy: 0.7952 - mean_pred: 0.3570 Epoch 611/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3932 - accuracy: 0.7952 - mean_pred: 0.3585 Epoch 612/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3902 - accuracy: 0.7989 - mean_pred: 0.3526 Epoch 613/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3905 - accuracy: 0.7952 - mean_pred: 0.3526 Epoch 614/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3933 - accuracy: 0.7877 - mean_pred: 0.3639 Epoch 615/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3907 - accuracy: 0.7952 - mean_pred: 0.3581 Epoch 616/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3921 - accuracy: 0.8007 - mean_pred: 0.3565 Epoch 617/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3921 - accuracy: 0.7914 - mean_pred: 0.3584 Epoch 618/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3903 - accuracy: 0.7952 - mean_pred: 0.3608 Epoch 619/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3900 - accuracy: 0.7970 - mean_pred: 0.3535 Epoch 620/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3906 - accuracy: 0.7933 - mean_pred: 0.3688 Epoch 621/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3912 - accuracy: 0.7989 - mean_pred: 0.3544 Epoch 622/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3903 - accuracy: 0.7914 - mean_pred: 0.3573 Epoch 623/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3898 - accuracy: 0.7989 - mean_pred: 0.3563 Epoch 624/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3903 - accuracy: 0.7858 - mean_pred: 0.3579 Epoch 625/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3922 - accuracy: 0.7914 - mean_pred: 0.3571 Epoch 626/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3913 - accuracy: 0.7877 - mean_pred: 0.3575 Epoch 627/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3923 - accuracy: 0.7970 - mean_pred: 0.3563 Epoch 628/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3889 - accuracy: 0.7914 - mean_pred: 0.3578 Epoch 629/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3910 - accuracy: 0.7952 - mean_pred: 0.3592 Epoch 630/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3887 - accuracy: 0.7989 - mean_pred: 0.3555 Epoch 631/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3901 - accuracy: 0.7933 - mean_pred: 0.3577 Epoch 632/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3908 - accuracy: 0.7858 - mean_pred: 0.3565 Epoch 633/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3889 - accuracy: 0.7896 - mean_pred: 0.3647 Epoch 634/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3900 - accuracy: 0.7989 - mean_pred: 0.3546 Epoch 635/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3890 - accuracy: 0.8007 - mean_pred: 0.3491 Epoch 636/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3891 - accuracy: 0.7933 - mean_pred: 0.3640 Epoch 637/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3900 - accuracy: 0.7877 - mean_pred: 0.3536 Epoch 638/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3911 - accuracy: 0.8045 - mean_pred: 0.3611 Epoch 639/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3893 - accuracy: 0.7970 - mean_pred: 0.3563 Epoch 640/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3895 - accuracy: 0.8026 - mean_pred: 0.3670 Epoch 641/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3882 - accuracy: 0.7933 - mean_pred: 0.3608 Epoch 642/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3882 - accuracy: 0.7989 - mean_pred: 0.3481 Epoch 643/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3902 - accuracy: 0.7914 - mean_pred: 0.3630 Epoch 644/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3888 - accuracy: 0.7914 - mean_pred: 0.3570 Epoch 645/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3889 - accuracy: 0.7933 - mean_pred: 0.3552 Epoch 646/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3882 - accuracy: 0.7952 - mean_pred: 0.3613 Epoch 647/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3890 - accuracy: 0.7896 - mean_pred: 0.3557 Epoch 648/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3884 - accuracy: 0.7952 - mean_pred: 0.3569 Epoch 649/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3885 - accuracy: 0.7970 - mean_pred: 0.3594 Epoch 650/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3880 - accuracy: 0.7952 - mean_pred: 0.3586 Epoch 651/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3887 - accuracy: 0.7933 - mean_pred: 0.3603 Epoch 652/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3869 - accuracy: 0.7989 - mean_pred: 0.3592 Epoch 653/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3897 - accuracy: 0.7858 - mean_pred: 0.3618 Epoch 654/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3885 - accuracy: 0.7989 - mean_pred: 0.3517 Epoch 655/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3873 - accuracy: 0.7933 - mean_pred: 0.3657 Epoch 656/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3876 - accuracy: 0.7970 - mean_pred: 0.3467 Epoch 657/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3873 - accuracy: 0.7952 - mean_pred: 0.3645 Epoch 658/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3864 - accuracy: 0.7896 - mean_pred: 0.3584 Epoch 659/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3908 - accuracy: 0.7877 - mean_pred: 0.3505 Epoch 660/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3877 - accuracy: 0.7896 - mean_pred: 0.3581 Epoch 661/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3855 - accuracy: 0.8007 - mean_pred: 0.3523 Epoch 662/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3895 - accuracy: 0.7933 - mean_pred: 0.3634 Epoch 663/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3893 - accuracy: 0.7933 - mean_pred: 0.3621 Epoch 664/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3860 - accuracy: 0.7896 - mean_pred: 0.3651 Epoch 665/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3892 - accuracy: 0.7952 - mean_pred: 0.3586 Epoch 666/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3866 - accuracy: 0.7933 - mean_pred: 0.3540 Epoch 667/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3857 - accuracy: 0.7970 - mean_pred: 0.3668 Epoch 668/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3870 - accuracy: 0.7952 - mean_pred: 0.3508 Epoch 669/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3892 - accuracy: 0.7877 - mean_pred: 0.3654 Epoch 670/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3853 - accuracy: 0.7933 - mean_pred: 0.3553 Epoch 671/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3858 - accuracy: 0.7896 - mean_pred: 0.3571 Epoch 672/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3864 - accuracy: 0.7914 - mean_pred: 0.3577 Epoch 673/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3851 - accuracy: 0.7858 - mean_pred: 0.3559 Epoch 674/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3871 - accuracy: 0.7933 - mean_pred: 0.3581 Epoch 675/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3866 - accuracy: 0.7896 - mean_pred: 0.3614 Epoch 676/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3881 - accuracy: 0.7933 - mean_pred: 0.3587 Epoch 677/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3854 - accuracy: 0.7896 - mean_pred: 0.3592 Epoch 678/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3861 - accuracy: 0.7970 - mean_pred: 0.3533 Epoch 679/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3852 - accuracy: 0.7877 - mean_pred: 0.3635 Epoch 680/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3860 - accuracy: 0.7858 - mean_pred: 0.3544 Epoch 681/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3854 - accuracy: 0.7914 - mean_pred: 0.3612 Epoch 682/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3862 - accuracy: 0.7914 - mean_pred: 0.3492 Epoch 683/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3858 - accuracy: 0.7896 - mean_pred: 0.3638 Epoch 684/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3852 - accuracy: 0.7970 - mean_pred: 0.3593 Epoch 685/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3874 - accuracy: 0.7914 - mean_pred: 0.3610 Epoch 686/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3846 - accuracy: 0.7933 - mean_pred: 0.3586 Epoch 687/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3852 - accuracy: 0.8007 - mean_pred: 0.3522 Epoch 688/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3845 - accuracy: 0.7914 - mean_pred: 0.3568 Epoch 689/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3847 - accuracy: 0.7933 - mean_pred: 0.3606 Epoch 690/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3852 - accuracy: 0.7877 - mean_pred: 0.3612 Epoch 691/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3848 - accuracy: 0.7952 - mean_pred: 0.3573 Epoch 692/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3860 - accuracy: 0.7933 - mean_pred: 0.3578 Epoch 693/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3840 - accuracy: 0.7952 - mean_pred: 0.3580 Epoch 694/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3837 - accuracy: 0.7933 - mean_pred: 0.3615 Epoch 695/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3839 - accuracy: 0.7970 - mean_pred: 0.3515 Epoch 696/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3822 - accuracy: 0.7989 - mean_pred: 0.3703 Epoch 697/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3848 - accuracy: 0.7952 - mean_pred: 0.3569 Epoch 698/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3853 - accuracy: 0.7877 - mean_pred: 0.3549 Epoch 699/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3826 - accuracy: 0.7970 - mean_pred: 0.3693 Epoch 700/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3858 - accuracy: 0.7877 - mean_pred: 0.3539 Epoch 701/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3846 - accuracy: 0.7970 - mean_pred: 0.3547 Epoch 702/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3824 - accuracy: 0.7970 - mean_pred: 0.3603 Epoch 703/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3847 - accuracy: 0.7952 - mean_pred: 0.3504 Epoch 704/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3813 - accuracy: 0.7933 - mean_pred: 0.3536 Epoch 705/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3843 - accuracy: 0.7933 - mean_pred: 0.3625 Epoch 706/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3833 - accuracy: 0.7989 - mean_pred: 0.3554 Epoch 707/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3832 - accuracy: 0.7914 - mean_pred: 0.3660 Epoch 708/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3827 - accuracy: 0.7858 - mean_pred: 0.3690 Epoch 709/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3827 - accuracy: 0.7952 - mean_pred: 0.3590 Epoch 710/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3820 - accuracy: 0.7970 - mean_pred: 0.3510 Epoch 711/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3822 - accuracy: 0.7952 - mean_pred: 0.3570 Epoch 712/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3827 - accuracy: 0.7989 - mean_pred: 0.3598 Epoch 713/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3815 - accuracy: 0.7952 - mean_pred: 0.3681 Epoch 714/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3836 - accuracy: 0.7952 - mean_pred: 0.3489 Epoch 715/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3808 - accuracy: 0.7933 - mean_pred: 0.3548 Epoch 716/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3854 - accuracy: 0.7952 - mean_pred: 0.3683 Epoch 717/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3800 - accuracy: 0.7877 - mean_pred: 0.3531 Epoch 718/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3811 - accuracy: 0.7952 - mean_pred: 0.3639 Epoch 719/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3814 - accuracy: 0.7914 - mean_pred: 0.3519 Epoch 720/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3811 - accuracy: 0.7952 - mean_pred: 0.3600 Epoch 721/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3814 - accuracy: 0.7914 - mean_pred: 0.3585 Epoch 722/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3808 - accuracy: 0.7970 - mean_pred: 0.3624 Epoch 723/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3814 - accuracy: 0.7989 - mean_pred: 0.3565 Epoch 724/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3798 - accuracy: 0.7933 - mean_pred: 0.3595 Epoch 725/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3820 - accuracy: 0.8007 - mean_pred: 0.3600 Epoch 726/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3796 - accuracy: 0.7896 - mean_pred: 0.3558 Epoch 727/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3802 - accuracy: 0.7877 - mean_pred: 0.3598 Epoch 728/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3802 - accuracy: 0.7933 - mean_pred: 0.3554 Epoch 729/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3805 - accuracy: 0.7896 - mean_pred: 0.3570 Epoch 730/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3801 - accuracy: 0.7933 - mean_pred: 0.3582 Epoch 731/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3799 - accuracy: 0.7914 - mean_pred: 0.3660 Epoch 732/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3810 - accuracy: 0.8007 - mean_pred: 0.3649 Epoch 733/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3789 - accuracy: 0.7952 - mean_pred: 0.3600 Epoch 734/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3796 - accuracy: 0.7933 - mean_pred: 0.3576 Epoch 735/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3795 - accuracy: 0.7952 - mean_pred: 0.3568 Epoch 736/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3813 - accuracy: 0.7952 - mean_pred: 0.3656 Epoch 737/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3782 - accuracy: 0.7896 - mean_pred: 0.3599 Epoch 738/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3791 - accuracy: 0.7877 - mean_pred: 0.3593 Epoch 739/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3789 - accuracy: 0.7970 - mean_pred: 0.3556 Epoch 740/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3801 - accuracy: 0.7933 - mean_pred: 0.3594 Epoch 741/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3779 - accuracy: 0.7896 - mean_pred: 0.3534 Epoch 742/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3800 - accuracy: 0.7877 - mean_pred: 0.3717 Epoch 743/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3784 - accuracy: 0.7914 - mean_pred: 0.3553 Epoch 744/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3773 - accuracy: 0.7933 - mean_pred: 0.3586 Epoch 745/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3792 - accuracy: 0.7952 - mean_pred: 0.3759 Epoch 746/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3790 - accuracy: 0.7858 - mean_pred: 0.3519 Epoch 747/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3784 - accuracy: 0.7914 - mean_pred: 0.3565 Epoch 748/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3792 - accuracy: 0.7952 - mean_pred: 0.3571 Epoch 749/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3780 - accuracy: 0.7914 - mean_pred: 0.3561 Epoch 750/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3782 - accuracy: 0.7952 - mean_pred: 0.3685 Epoch 751/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3790 - accuracy: 0.7877 - mean_pred: 0.3551 Epoch 752/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3777 - accuracy: 0.7896 - mean_pred: 0.3600 Epoch 753/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3767 - accuracy: 0.7914 - mean_pred: 0.3602 Epoch 754/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3786 - accuracy: 0.7933 - mean_pred: 0.3559 Epoch 755/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3781 - accuracy: 0.7896 - mean_pred: 0.3652 Epoch 756/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3765 - accuracy: 0.8101 - mean_pred: 0.3490 Epoch 757/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3781 - accuracy: 0.7933 - mean_pred: 0.3595 Epoch 758/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3771 - accuracy: 0.7970 - mean_pred: 0.3707 Epoch 759/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3774 - accuracy: 0.7877 - mean_pred: 0.3498 Epoch 760/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3756 - accuracy: 0.7877 - mean_pred: 0.3648 Epoch 761/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3780 - accuracy: 0.7952 - mean_pred: 0.3504 Epoch 762/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3763 - accuracy: 0.7933 - mean_pred: 0.3641 Epoch 763/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3760 - accuracy: 0.7840 - mean_pred: 0.3600 Epoch 764/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3766 - accuracy: 0.7914 - mean_pred: 0.3585 Epoch 765/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3766 - accuracy: 0.7952 - mean_pred: 0.3450 Epoch 766/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3768 - accuracy: 0.7970 - mean_pred: 0.3698 Epoch 767/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3760 - accuracy: 0.7877 - mean_pred: 0.3569 Epoch 768/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3753 - accuracy: 0.7952 - mean_pred: 0.3573 Epoch 769/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3749 - accuracy: 0.7989 - mean_pred: 0.3569 Epoch 770/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3757 - accuracy: 0.8007 - mean_pred: 0.3648 Epoch 771/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3750 - accuracy: 0.7896 - mean_pred: 0.3587 Epoch 772/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3748 - accuracy: 0.7933 - mean_pred: 0.3604 Epoch 773/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3752 - accuracy: 0.7858 - mean_pred: 0.3604 Epoch 774/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3760 - accuracy: 0.7933 - mean_pred: 0.3645 Epoch 775/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3735 - accuracy: 0.7896 - mean_pred: 0.3588 Epoch 776/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3749 - accuracy: 0.7933 - mean_pred: 0.3651 Epoch 777/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3756 - accuracy: 0.7933 - mean_pred: 0.3573 Epoch 778/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3742 - accuracy: 0.8007 - mean_pred: 0.3555 Epoch 779/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3736 - accuracy: 0.7933 - mean_pred: 0.3617 Epoch 780/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3743 - accuracy: 0.7896 - mean_pred: 0.3563 Epoch 781/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3734 - accuracy: 0.7914 - mean_pred: 0.3574 Epoch 782/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3731 - accuracy: 0.8007 - mean_pred: 0.3604 Epoch 783/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3753 - accuracy: 0.7914 - mean_pred: 0.3589 Epoch 784/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3739 - accuracy: 0.7933 - mean_pred: 0.3555 Epoch 785/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3732 - accuracy: 0.7970 - mean_pred: 0.3620 Epoch 786/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3729 - accuracy: 0.7970 - mean_pred: 0.3579 Epoch 787/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3733 - accuracy: 0.8007 - mean_pred: 0.3644 Epoch 788/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3714 - accuracy: 0.8007 - mean_pred: 0.3537 Epoch 789/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3718 - accuracy: 0.8007 - mean_pred: 0.3605 Epoch 790/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3727 - accuracy: 0.7933 - mean_pred: 0.3623 Epoch 791/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3718 - accuracy: 0.7970 - mean_pred: 0.3600 Epoch 792/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3729 - accuracy: 0.7970 - mean_pred: 0.3562 Epoch 793/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3712 - accuracy: 0.8007 - mean_pred: 0.3527 Epoch 794/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3732 - accuracy: 0.7914 - mean_pred: 0.3645 Epoch 795/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3718 - accuracy: 0.7989 - mean_pred: 0.3636 Epoch 796/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3695 - accuracy: 0.8045 - mean_pred: 0.3529 Epoch 797/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3714 - accuracy: 0.7952 - mean_pred: 0.3705 Epoch 798/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3716 - accuracy: 0.7970 - mean_pred: 0.3602 Epoch 799/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3714 - accuracy: 0.7952 - mean_pred: 0.3621 Epoch 800/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3705 - accuracy: 0.8101 - mean_pred: 0.3523 Epoch 801/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3698 - accuracy: 0.7952 - mean_pred: 0.3647 Epoch 802/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3699 - accuracy: 0.7952 - mean_pred: 0.3491 Epoch 803/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3717 - accuracy: 0.7970 - mean_pred: 0.3635 Epoch 804/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3692 - accuracy: 0.8026 - mean_pred: 0.3705 Epoch 805/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3704 - accuracy: 0.8026 - mean_pred: 0.3506 Epoch 806/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3701 - accuracy: 0.7914 - mean_pred: 0.3577 Epoch 807/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3698 - accuracy: 0.7970 - mean_pred: 0.3583 Epoch 808/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3701 - accuracy: 0.7896 - mean_pred: 0.3582 Epoch 809/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3713 - accuracy: 0.7952 - mean_pred: 0.3565 Epoch 810/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3692 - accuracy: 0.7933 - mean_pred: 0.3638 Epoch 811/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3708 - accuracy: 0.7989 - mean_pred: 0.3562 Epoch 812/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3706 - accuracy: 0.7933 - mean_pred: 0.3584 Epoch 813/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3688 - accuracy: 0.8026 - mean_pred: 0.3635 Epoch 814/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3691 - accuracy: 0.7989 - mean_pred: 0.3611 Epoch 815/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3709 - accuracy: 0.7896 - mean_pred: 0.3575 Epoch 816/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3707 - accuracy: 0.8007 - mean_pred: 0.3559 Epoch 817/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3696 - accuracy: 0.7989 - mean_pred: 0.3462 Epoch 818/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3712 - accuracy: 0.7933 - mean_pred: 0.3631 Epoch 819/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3696 - accuracy: 0.7989 - mean_pred: 0.3610 Epoch 820/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3681 - accuracy: 0.7989 - mean_pred: 0.3572 Epoch 821/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3686 - accuracy: 0.8082 - mean_pred: 0.3644 Epoch 822/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3673 - accuracy: 0.7989 - mean_pred: 0.3556 Epoch 823/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3682 - accuracy: 0.7952 - mean_pred: 0.3573 Epoch 824/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3698 - accuracy: 0.7989 - mean_pred: 0.3679 Epoch 825/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3683 - accuracy: 0.7933 - mean_pred: 0.3583 Epoch 826/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3665 - accuracy: 0.7952 - mean_pred: 0.3572 Epoch 827/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3679 - accuracy: 0.7952 - mean_pred: 0.3787 Epoch 828/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3672 - accuracy: 0.8007 - mean_pred: 0.3495 Epoch 829/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3673 - accuracy: 0.7970 - mean_pred: 0.3674 Epoch 830/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3689 - accuracy: 0.8045 - mean_pred: 0.3454 Epoch 831/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3670 - accuracy: 0.7970 - mean_pred: 0.3660 Epoch 832/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3681 - accuracy: 0.8082 - mean_pred: 0.3576 Epoch 833/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3671 - accuracy: 0.7933 - mean_pred: 0.3584 Epoch 834/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3671 - accuracy: 0.8101 - mean_pred: 0.3629 Epoch 835/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3694 - accuracy: 0.7970 - mean_pred: 0.3575 Epoch 836/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3660 - accuracy: 0.8007 - mean_pred: 0.3512 Epoch 837/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3662 - accuracy: 0.8026 - mean_pred: 0.3573 Epoch 838/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3672 - accuracy: 0.7989 - mean_pred: 0.3533 Epoch 839/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3661 - accuracy: 0.7989 - mean_pred: 0.3624 Epoch 840/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3674 - accuracy: 0.7896 - mean_pred: 0.3586 Epoch 841/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3661 - accuracy: 0.7989 - mean_pred: 0.3584 Epoch 842/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3672 - accuracy: 0.7970 - mean_pred: 0.3629 Epoch 843/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3653 - accuracy: 0.8026 - mean_pred: 0.3609 Epoch 844/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3651 - accuracy: 0.8045 - mean_pred: 0.3670 Epoch 845/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3662 - accuracy: 0.8045 - mean_pred: 0.3538 Epoch 846/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3661 - accuracy: 0.8063 - mean_pred: 0.3614 Epoch 847/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3652 - accuracy: 0.8026 - mean_pred: 0.3555 Epoch 848/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3660 - accuracy: 0.7989 - mean_pred: 0.3590 Epoch 849/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3660 - accuracy: 0.8026 - mean_pred: 0.3582 Epoch 850/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3648 - accuracy: 0.8063 - mean_pred: 0.3557 Epoch 851/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3647 - accuracy: 0.8026 - mean_pred: 0.3640 Epoch 852/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3649 - accuracy: 0.8138 - mean_pred: 0.3596 Epoch 853/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3658 - accuracy: 0.7970 - mean_pred: 0.3499 Epoch 854/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3652 - accuracy: 0.8026 - mean_pred: 0.3581 Epoch 855/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3638 - accuracy: 0.7989 - mean_pred: 0.3588 Epoch 856/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3640 - accuracy: 0.7970 - mean_pred: 0.3549 Epoch 857/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3653 - accuracy: 0.7989 - mean_pred: 0.3689 Epoch 858/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3663 - accuracy: 0.8026 - mean_pred: 0.3628 Epoch 859/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3638 - accuracy: 0.8026 - mean_pred: 0.3562 Epoch 860/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3633 - accuracy: 0.8007 - mean_pred: 0.3576 Epoch 861/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3628 - accuracy: 0.8045 - mean_pred: 0.3594 Epoch 862/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3654 - accuracy: 0.7989 - mean_pred: 0.3503 Epoch 863/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3653 - accuracy: 0.7989 - mean_pred: 0.3567 Epoch 864/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3632 - accuracy: 0.7970 - mean_pred: 0.3663 Epoch 865/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3644 - accuracy: 0.7952 - mean_pred: 0.3587 Epoch 866/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3622 - accuracy: 0.8138 - mean_pred: 0.3459 Epoch 867/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3635 - accuracy: 0.7970 - mean_pred: 0.3594 Epoch 868/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3640 - accuracy: 0.7989 - mean_pred: 0.3618 Epoch 869/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3615 - accuracy: 0.8026 - mean_pred: 0.3546 Epoch 870/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3639 - accuracy: 0.7933 - mean_pred: 0.3599 Epoch 871/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3626 - accuracy: 0.7933 - mean_pred: 0.3610 Epoch 872/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3629 - accuracy: 0.8063 - mean_pred: 0.3649 Epoch 873/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3622 - accuracy: 0.8063 - mean_pred: 0.3523 Epoch 874/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3643 - accuracy: 0.7952 - mean_pred: 0.3658 Epoch 875/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3614 - accuracy: 0.7952 - mean_pred: 0.3546 Epoch 876/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3631 - accuracy: 0.7952 - mean_pred: 0.3647 Epoch 877/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3619 - accuracy: 0.8026 - mean_pred: 0.3575 Epoch 878/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3622 - accuracy: 0.8045 - mean_pred: 0.3607 Epoch 879/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3639 - accuracy: 0.8063 - mean_pred: 0.3590 Epoch 880/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3617 - accuracy: 0.7989 - mean_pred: 0.3549 Epoch 881/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3615 - accuracy: 0.8026 - mean_pred: 0.3591 Epoch 882/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3610 - accuracy: 0.7989 - mean_pred: 0.3603 Epoch 883/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3626 - accuracy: 0.8045 - mean_pred: 0.3549 Epoch 884/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3609 - accuracy: 0.8007 - mean_pred: 0.3553 Epoch 885/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3599 - accuracy: 0.8007 - mean_pred: 0.3698 Epoch 886/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3606 - accuracy: 0.8026 - mean_pred: 0.3572 Epoch 887/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3613 - accuracy: 0.7989 - mean_pred: 0.3474 Epoch 888/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3602 - accuracy: 0.8026 - mean_pred: 0.3639 Epoch 889/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3603 - accuracy: 0.8082 - mean_pred: 0.3552 Epoch 890/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3620 - accuracy: 0.8045 - mean_pred: 0.3595 Epoch 891/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3613 - accuracy: 0.8007 - mean_pred: 0.3521 Epoch 892/1000 537/537 [==============================] - 0s 13us/step - loss: 0.3599 - accuracy: 0.8045 - mean_pred: 0.3639 Epoch 893/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3590 - accuracy: 0.8101 - mean_pred: 0.3547 Epoch 894/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3612 - accuracy: 0.7989 - mean_pred: 0.3627 Epoch 895/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3605 - accuracy: 0.8007 - mean_pred: 0.3574 Epoch 896/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3605 - accuracy: 0.8045 - mean_pred: 0.3555 Epoch 897/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3615 - accuracy: 0.8007 - mean_pred: 0.3634 Epoch 898/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3586 - accuracy: 0.8082 - mean_pred: 0.3481 Epoch 899/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3612 - accuracy: 0.7970 - mean_pred: 0.3687 Epoch 900/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3584 - accuracy: 0.7989 - mean_pred: 0.3462 Epoch 901/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3572 - accuracy: 0.7933 - mean_pred: 0.3628 Epoch 902/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3597 - accuracy: 0.8045 - mean_pred: 0.3611 Epoch 903/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3592 - accuracy: 0.7952 - mean_pred: 0.3628 Epoch 904/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3566 - accuracy: 0.8007 - mean_pred: 0.3526 Epoch 905/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3611 - accuracy: 0.8045 - mean_pred: 0.3615 Epoch 906/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3602 - accuracy: 0.7989 - mean_pred: 0.3562 Epoch 907/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3595 - accuracy: 0.8045 - mean_pred: 0.3641 Epoch 908/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3574 - accuracy: 0.8045 - mean_pred: 0.3497 Epoch 909/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3588 - accuracy: 0.8007 - mean_pred: 0.3648 Epoch 910/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3572 - accuracy: 0.8007 - mean_pred: 0.3527 Epoch 911/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3580 - accuracy: 0.7952 - mean_pred: 0.3591 Epoch 912/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3597 - accuracy: 0.7970 - mean_pred: 0.3608 Epoch 913/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3565 - accuracy: 0.8063 - mean_pred: 0.3517 Epoch 914/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3599 - accuracy: 0.7989 - mean_pred: 0.3631 Epoch 915/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3575 - accuracy: 0.8045 - mean_pred: 0.3625 Epoch 916/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3580 - accuracy: 0.7970 - mean_pred: 0.3616 Epoch 917/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3559 - accuracy: 0.8063 - mean_pred: 0.3660 Epoch 918/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3587 - accuracy: 0.8007 - mean_pred: 0.3541 Epoch 919/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3561 - accuracy: 0.8026 - mean_pred: 0.3575 Epoch 920/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3569 - accuracy: 0.8063 - mean_pred: 0.3535 Epoch 921/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3590 - accuracy: 0.8082 - mean_pred: 0.3681 Epoch 922/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3543 - accuracy: 0.8026 - mean_pred: 0.3548 Epoch 923/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3568 - accuracy: 0.8007 - mean_pred: 0.3549 Epoch 924/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3573 - accuracy: 0.7970 - mean_pred: 0.3657 Epoch 925/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3552 - accuracy: 0.8063 - mean_pred: 0.3617 Epoch 926/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3578 - accuracy: 0.7970 - mean_pred: 0.3548 Epoch 927/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3544 - accuracy: 0.8063 - mean_pred: 0.3585 Epoch 928/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3564 - accuracy: 0.8007 - mean_pred: 0.3557 Epoch 929/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3557 - accuracy: 0.7952 - mean_pred: 0.3629 Epoch 930/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3554 - accuracy: 0.7970 - mean_pred: 0.3527 Epoch 931/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3560 - accuracy: 0.8026 - mean_pred: 0.3548 Epoch 932/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3546 - accuracy: 0.8026 - mean_pred: 0.3632 Epoch 933/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3573 - accuracy: 0.7970 - mean_pred: 0.3595 Epoch 934/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3553 - accuracy: 0.8082 - mean_pred: 0.3593 Epoch 935/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3556 - accuracy: 0.7989 - mean_pred: 0.3607 Epoch 936/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3536 - accuracy: 0.8082 - mean_pred: 0.3576 Epoch 937/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3549 - accuracy: 0.8026 - mean_pred: 0.3711 Epoch 938/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3548 - accuracy: 0.8045 - mean_pred: 0.3535 Epoch 939/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3541 - accuracy: 0.8063 - mean_pred: 0.3594 Epoch 940/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3526 - accuracy: 0.8026 - mean_pred: 0.3645 Epoch 941/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3543 - accuracy: 0.8026 - mean_pred: 0.3513 Epoch 942/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3552 - accuracy: 0.8082 - mean_pred: 0.3643 Epoch 943/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3522 - accuracy: 0.8063 - mean_pred: 0.3555 Epoch 944/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3526 - accuracy: 0.8138 - mean_pred: 0.3551 Epoch 945/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3529 - accuracy: 0.8082 - mean_pred: 0.3699 Epoch 946/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3549 - accuracy: 0.8138 - mean_pred: 0.3558 Epoch 947/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3530 - accuracy: 0.7970 - mean_pred: 0.3609 Epoch 948/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3540 - accuracy: 0.7970 - mean_pred: 0.3628 Epoch 949/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3536 - accuracy: 0.8082 - mean_pred: 0.3527 Epoch 950/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3544 - accuracy: 0.8026 - mean_pred: 0.3605 Epoch 951/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3529 - accuracy: 0.8119 - mean_pred: 0.3618 Epoch 952/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3521 - accuracy: 0.8063 - mean_pred: 0.3659 Epoch 953/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3543 - accuracy: 0.8082 - mean_pred: 0.3510 Epoch 954/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3535 - accuracy: 0.8063 - mean_pred: 0.3574 Epoch 955/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3516 - accuracy: 0.8063 - mean_pred: 0.3637 Epoch 956/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3514 - accuracy: 0.8119 - mean_pred: 0.3653 Epoch 957/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3528 - accuracy: 0.8045 - mean_pred: 0.3552 Epoch 958/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3530 - accuracy: 0.8026 - mean_pred: 0.3650 Epoch 959/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3519 - accuracy: 0.8119 - mean_pred: 0.3594 Epoch 960/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3513 - accuracy: 0.8082 - mean_pred: 0.3561 Epoch 961/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3506 - accuracy: 0.8045 - mean_pred: 0.3593 Epoch 962/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3522 - accuracy: 0.8082 - mean_pred: 0.3779 Epoch 963/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3505 - accuracy: 0.8156 - mean_pred: 0.3525 Epoch 964/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3517 - accuracy: 0.8082 - mean_pred: 0.3671 Epoch 965/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3512 - accuracy: 0.8119 - mean_pred: 0.3589 Epoch 966/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3512 - accuracy: 0.8063 - mean_pred: 0.3658 Epoch 967/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3511 - accuracy: 0.8101 - mean_pred: 0.3576 Epoch 968/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3509 - accuracy: 0.8101 - mean_pred: 0.3617 Epoch 969/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3504 - accuracy: 0.8026 - mean_pred: 0.3718 Epoch 970/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3524 - accuracy: 0.8026 - mean_pred: 0.3596 Epoch 971/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3505 - accuracy: 0.8101 - mean_pred: 0.3615 Epoch 972/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3503 - accuracy: 0.8101 - mean_pred: 0.3659 Epoch 973/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3505 - accuracy: 0.8138 - mean_pred: 0.3479 Epoch 974/1000 537/537 [==============================] - 0s 19us/step - loss: 0.3517 - accuracy: 0.8045 - mean_pred: 0.3609 Epoch 975/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3500 - accuracy: 0.8007 - mean_pred: 0.3578 Epoch 976/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3498 - accuracy: 0.8063 - mean_pred: 0.3626 Epoch 977/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3483 - accuracy: 0.8101 - mean_pred: 0.3678 Epoch 978/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3513 - accuracy: 0.8138 - mean_pred: 0.3520 Epoch 979/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3523 - accuracy: 0.8082 - mean_pred: 0.3631 Epoch 980/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3475 - accuracy: 0.8082 - mean_pred: 0.3621 Epoch 981/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3501 - accuracy: 0.8063 - mean_pred: 0.3621 Epoch 982/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3493 - accuracy: 0.8101 - mean_pred: 0.3598 Epoch 983/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3509 - accuracy: 0.8082 - mean_pred: 0.3759 Epoch 984/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3482 - accuracy: 0.8082 - mean_pred: 0.3519 Epoch 985/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3501 - accuracy: 0.8026 - mean_pred: 0.3657 Epoch 986/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3470 - accuracy: 0.8156 - mean_pred: 0.3608 Epoch 987/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3517 - accuracy: 0.8045 - mean_pred: 0.3647 Epoch 988/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3495 - accuracy: 0.8082 - mean_pred: 0.3540 Epoch 989/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3481 - accuracy: 0.8082 - mean_pred: 0.3690 Epoch 990/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3486 - accuracy: 0.8119 - mean_pred: 0.3503 Epoch 991/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3492 - accuracy: 0.8138 - mean_pred: 0.3665 Epoch 992/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3484 - accuracy: 0.8045 - mean_pred: 0.3651 Epoch 993/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3494 - accuracy: 0.8026 - mean_pred: 0.3626 Epoch 994/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3483 - accuracy: 0.8101 - mean_pred: 0.3631 Epoch 995/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3472 - accuracy: 0.8175 - mean_pred: 0.3658 Epoch 996/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3473 - accuracy: 0.8082 - mean_pred: 0.3585 Epoch 997/1000 537/537 [==============================] - 0s 15us/step - loss: 0.3492 - accuracy: 0.8156 - mean_pred: 0.3621 Epoch 998/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3460 - accuracy: 0.8119 - mean_pred: 0.3769 Epoch 999/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3494 - accuracy: 0.8101 - mean_pred: 0.3550 Epoch 1000/1000 537/537 [==============================] - 0s 17us/step - loss: 0.3458 - accuracy: 0.8119 - mean_pred: 0.3574 231/231 [==============================] - 0s 82us/step
In [13]:
model.summary()
Model: "sequential_1" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= dense_1 (Dense) (None, 12) 108 _________________________________________________________________ dense_2 (Dense) (None, 10) 130 _________________________________________________________________ dense_3 (Dense) (None, 1) 11 ================================================================= Total params: 249 Trainable params: 249 Non-trainable params: 0 _________________________________________________________________
Accuracy¶
In [14]:
score = pd.DataFrame(score, index = model.metrics_names).T
history = pd.DataFrame(history.history)
display(score.style.hide_index())
| loss | accuracy | mean_pred |
|---|---|---|
| 1.284564 | 0.735931 | 0.430036 |
In [15]:
fig, ax = plt.subplots(1, 1, figsize=(12, 6))
_ = ax.plot(history['accuracy'], 'navy', label='Accuracy', linewidth=2)
_ = ax.plot(history['loss'], 'red', label='Loss', linewidth=2)
_ = ax.set_yscale('log')
_ = ax.set_xlim(left = 0, right = N)
_ = ax.legend(bbox_to_anchor=(1.05, 1), loc='upper left', borderaxespad=0., fontsize = 14)
_ = ax.set_xlabel('Steps', fontsize = 14)
As expected, the accuracy and loss improve as step number increases.
A Graph of the Model¶
In [16]:
plot_model(model, show_shapes=True, show_layer_names=True, expand_nested = True)
Out[16]:
